Commit 86f7f9c4 authored by Alexander Henkel's avatar Alexander Henkel
Browse files

bugfixing

parent 0af01b36
%% Cell type:code id:18d33b57 tags:
 
``` python
%load_ext autoreload
%autoreload 2
 
%matplotlib notebook
```
 
%% Cell type:code id:5b007717 tags:
 
``` python
import sys
import os
import numpy as np
import matplotlib.pyplot as plt
from IPython.display import display, Markdown
import copy
import pandas as pd
```
 
%% Cell type:code id:c32112e2 tags:
 
``` python
module_path = os.path.abspath(os.path.join('..'))
os.chdir(module_path)
if module_path not in sys.path:
sys.path.append(module_path)
```
 
%% Cell type:code id:b0e20ed4 tags:
 
``` python
from personalization_tools.load_data_sets import *
from personalization_tools.helpers import *
from personalization_tools.learner_pipeline import LearnerPipeline, Evaluation
from personalization_tools.dataset_builder import *
from personalization_tools.model_evaluation import ModelEvaluation
from personalization_tools.sensor_recorder_data_reader import SensorRecorderDataReader
from personalization_tools.dataset_manager import DatasetManager
from personalization_tools.trainings_manager import TrainingsManager
from personalization_tools.pseudo_model_settings import pseudo_model_settings, thesis_filters, translate_setting
from personalization_tools.pseudo_label_evaluation import *
from personalization_tools.globals import Indicators
from personalization_tools.evaluation_manager import EvaluationManager
```
 
%% Cell type:code id:db3ce6a9 tags:
 
``` python
pd.set_option('display.max_rows', 500)
```
 
%% Cell type:code id:2bdc9c75 tags:
 
``` python
evaluation_config_file = './data/cluster/pseudo_collections/evaluation_config.yaml'
```
 
%% Cell type:code id:c977159b tags:
 
``` python
evaluation_manager = EvaluationManager()
evaluation_manager.load_config(evaluation_config_file)
```
 
%% Output
 
load config
 
%% Cell type:code id:2f20d006 tags:
%% Cell type:code id:5f34a36f tags:
 
``` python
for model in evaluation_manager.get_run_of_collection('random_synthetic_01'):
evaluation_manager.model_evaluation.clear_evaluations_of_model(model)
```
 
%% Cell type:code id:6decca8f tags:
 
``` python
evaluation_manager.do_predictions()
```
 
%% Output
 
Do Predictions for:
Use device: cuda
run: ('random_01_run1_all_rn2ry0.pt', 01_generated_0) ... 0.8635909557342529 seconds
run: ('random_01_run1_all_rn2ry0.pt', 01_generated_1) ... 0.7559754848480225 seconds
run: ('random_01_run1_noneut_rn2ry0.pt', 01_generated_0) ... 0.769207239151001 seconds
run: ('random_01_run1_noneut_rn2ry0.pt', 01_generated_1) ... 0.7538094520568848 seconds
run: ('random_01_run1_high_rn2ry0.pt', 01_generated_0) ... 0.7616837024688721 seconds
run: ('random_01_run1_high_rn2ry0.pt', 01_generated_1) ... 0.7572121620178223 seconds
run: ('random_01_run1_all_corrected_rn2ry0.pt', 01_generated_0) ... 0.7925519943237305 seconds
run: ('random_01_run1_all_corrected_rn2ry0.pt', 01_generated_1) ... 0.7952256202697754 seconds
run: ('random_01_run1_all_corrected_noise_rn2ry0.pt', 01_generated_0) ... 0.7728948593139648 seconds
run: ('random_01_run1_all_corrected_noise_rn2ry0.pt', 01_generated_1) ... 0.7352817058563232 seconds
run: ('random_01_run1_scope_corrected_noise_rn2ry0.pt', 01_generated_0) ... 0.7921650409698486 seconds
run: ('random_01_run1_scope_corrected_noise_rn2ry0.pt', 01_generated_1) ... 0.7624926567077637 seconds
run: ('random_01_run1_noneut_corrected_rn2ry0.pt', 01_generated_0) ... 0.7853000164031982 seconds
run: ('random_01_run1_noneut_corrected_rn2ry0.pt', 01_generated_1) ... 0.7344820499420166 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn2ry0.pt', 01_generated_0) ... 0.8096168041229248 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn2ry0.pt', 01_generated_1) ... 0.7759416103363037 seconds
run: ('random_01_run1_allnoise_correctedscore_rn2ry0.pt', 01_generated_0) ... 0.7606852054595947 seconds
run: ('random_01_run1_allnoise_correctedscore_rn2ry0.pt', 01_generated_1) ... 0.7505133152008057 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn2ry0.pt', 01_generated_0) ... 0.7556631565093994 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn2ry0.pt', 01_generated_1) ... 0.7928900718688965 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn2ry0.pt', 01_generated_0) ... 0.7585153579711914 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn2ry0.pt', 01_generated_1) ... 0.7369773387908936 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn2ry0.pt', 01_generated_0) ... 0.755394697189331 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn2ry0.pt', 01_generated_1) ... 0.7497127056121826 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn2ry0.pt', 01_generated_0) ... 0.7556638717651367 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn2ry0.pt', 01_generated_1) ... 0.7547783851623535 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn2ry0.pt', 01_generated_0) ... 0.7777938842773438 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn2ry0.pt', 01_generated_1) ... 0.7589237689971924 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn2ry0.pt', 01_generated_0) ... 0.7642219066619873 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn2ry0.pt', 01_generated_1) ... 0.7259056568145752 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn2ry0.pt', 01_generated_0) ... 0.7517473697662354 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn2ry0.pt', 01_generated_1) ... 0.7441573143005371 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn2ry0.pt', 01_generated_0) ... 0.8003270626068115 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn2ry0.pt', 01_generated_1) ... 0.7511992454528809 seconds
run: ('random_01_run1_all_rn3ry0.pt', 01_generated_0) ... 0.7873153686523438 seconds
run: ('random_01_run1_all_rn3ry0.pt', 01_generated_1) ... 0.7366888523101807 seconds
run: ('random_01_run1_noneut_rn3ry0.pt', 01_generated_0) ... 0.7884097099304199 seconds
run: ('random_01_run1_noneut_rn3ry0.pt', 01_generated_1) ... 0.7301690578460693 seconds
run: ('random_01_run1_high_rn3ry0.pt', 01_generated_0) ... 0.754291296005249 seconds
run: ('random_01_run1_high_rn3ry0.pt', 01_generated_1) ... 0.7309882640838623 seconds
run: ('random_01_run1_all_corrected_rn3ry0.pt', 01_generated_0) ... 0.7821619510650635 seconds
run: ('random_01_run1_all_corrected_rn3ry0.pt', 01_generated_1) ... 0.7578375339508057 seconds
run: ('random_01_run1_all_corrected_noise_rn3ry0.pt', 01_generated_0) ... 0.7645065784454346 seconds
run: ('random_01_run1_all_corrected_noise_rn3ry0.pt', 01_generated_1) ... 0.7596299648284912 seconds
run: ('random_01_run1_scope_corrected_noise_rn3ry0.pt', 01_generated_0) ... 0.7941877841949463 seconds
run: ('random_01_run1_scope_corrected_noise_rn3ry0.pt', 01_generated_1) ... 0.7285556793212891 seconds
run: ('random_01_run1_noneut_corrected_rn3ry0.pt', 01_generated_0) ... 0.780742883682251 seconds
run: ('random_01_run1_noneut_corrected_rn3ry0.pt', 01_generated_1) ... 0.7276768684387207 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn3ry0.pt', 01_generated_0) ... 0.7688539028167725 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn3ry0.pt', 01_generated_1) ... 0.732712984085083 seconds
run: ('random_01_run1_allnoise_correctedscore_rn3ry0.pt', 01_generated_0) ... 0.7582430839538574 seconds
run: ('random_01_run1_allnoise_correctedscore_rn3ry0.pt', 01_generated_1) ... 0.7498233318328857 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn3ry0.pt', 01_generated_0) ... 0.8055596351623535 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn3ry0.pt', 01_generated_1) ... 0.8094184398651123 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn3ry0.pt', 01_generated_0) ... 0.8268589973449707 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn3ry0.pt', 01_generated_1) ... 0.7235498428344727 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn3ry0.pt', 01_generated_0) ... 0.8530466556549072 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn3ry0.pt', 01_generated_1) ... 0.8100605010986328 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn3ry0.pt', 01_generated_0) ... 0.8334698677062988 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn3ry0.pt', 01_generated_1) ... 0.7387728691101074 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn3ry0.pt', 01_generated_0) ... 0.8335132598876953 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn3ry0.pt', 01_generated_1) ... 0.8061039447784424 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn3ry0.pt', 01_generated_0) ... 0.8185348510742188 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn3ry0.pt', 01_generated_1) ... 0.7325232028961182 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn3ry0.pt', 01_generated_0) ... 0.8085942268371582 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn3ry0.pt', 01_generated_1) ... 0.7617881298065186 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn3ry0.pt', 01_generated_0) ... 0.8054249286651611 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn3ry0.pt', 01_generated_1) ... 0.7698974609375 seconds
run: ('random_01_run1_all_rn4ry0.pt', 01_generated_0) ... 0.8151412010192871 seconds
run: ('random_01_run1_all_rn4ry0.pt', 01_generated_1) ... 0.7857952117919922 seconds
run: ('random_01_run1_noneut_rn4ry0.pt', 01_generated_0) ... 0.7914161682128906 seconds
run: ('random_01_run1_noneut_rn4ry0.pt', 01_generated_1) ... 0.7642643451690674 seconds
run: ('random_01_run1_high_rn4ry0.pt', 01_generated_0) ... 0.8170721530914307 seconds
run: ('random_01_run1_high_rn4ry0.pt', 01_generated_1) ... 0.8080852031707764 seconds
run: ('random_01_run1_all_corrected_rn4ry0.pt', 01_generated_0) ... 0.8258817195892334 seconds
run: ('random_01_run1_all_corrected_rn4ry0.pt', 01_generated_1) ... 0.7757227420806885 seconds
run: ('random_01_run1_all_corrected_noise_rn4ry0.pt', 01_generated_0) ... 0.8063132762908936 seconds
run: ('random_01_run1_all_corrected_noise_rn4ry0.pt', 01_generated_1) ... 0.7625775337219238 seconds
 
run: ('random_01_run1_scope_corrected_noise_rn4ry0.pt', 01_generated_0) ... 0.7983028888702393 seconds
run: ('random_01_run1_scope_corrected_noise_rn4ry0.pt', 01_generated_1) ... 0.7788314819335938 seconds
run: ('random_01_run1_noneut_corrected_rn4ry0.pt', 01_generated_0) ... 0.833223819732666 seconds
run: ('random_01_run1_noneut_corrected_rn4ry0.pt', 01_generated_1) ... 0.7816512584686279 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn4ry0.pt', 01_generated_0) ... 0.7902383804321289 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn4ry0.pt', 01_generated_1) ... 0.735945463180542 seconds
run: ('random_01_run1_allnoise_correctedscore_rn4ry0.pt', 01_generated_0) ... 0.8140208721160889 seconds
run: ('random_01_run1_allnoise_correctedscore_rn4ry0.pt', 01_generated_1) ... 0.7781250476837158 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn4ry0.pt', 01_generated_0) ... 0.828782320022583 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn4ry0.pt', 01_generated_1) ... 0.7657344341278076 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn4ry0.pt', 01_generated_0) ... 0.8204920291900635 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn4ry0.pt', 01_generated_1) ... 0.7816641330718994 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn4ry0.pt', 01_generated_0) ... 0.7875323295593262 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn4ry0.pt', 01_generated_1) ... 0.7537124156951904 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn4ry0.pt', 01_generated_0) ... 0.8360104560852051 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn4ry0.pt', 01_generated_1) ... 0.802016019821167 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn4ry0.pt', 01_generated_0) ... 0.8066842555999756 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn4ry0.pt', 01_generated_1) ... 0.7857615947723389 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn4ry0.pt', 01_generated_0) ... 0.849196195602417 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn4ry0.pt', 01_generated_1) ... 0.8132874965667725 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn4ry0.pt', 01_generated_0) ... 0.7819747924804688 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn4ry0.pt', 01_generated_1) ... 0.7834398746490479 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn4ry0.pt', 01_generated_0) ... 0.8089962005615234 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn4ry0.pt', 01_generated_1) ... 0.7883977890014648 seconds
run: ('random_01_run1_all_rn5ry0.pt', 01_generated_0) ... 0.7895948886871338 seconds
run: ('random_01_run1_all_rn5ry0.pt', 01_generated_1) ... 0.7581923007965088 seconds
run: ('random_01_run1_noneut_rn5ry0.pt', 01_generated_0) ... 0.8186430931091309 seconds
run: ('random_01_run1_noneut_rn5ry0.pt', 01_generated_1) ... 0.7803764343261719 seconds
run: ('random_01_run1_high_rn5ry0.pt', 01_generated_0) ... 0.7627100944519043 seconds
run: ('random_01_run1_high_rn5ry0.pt', 01_generated_1) ... 0.7943999767303467 seconds
run: ('random_01_run1_all_corrected_rn5ry0.pt', 01_generated_0) ... 0.8065156936645508 seconds
run: ('random_01_run1_all_corrected_rn5ry0.pt', 01_generated_1) ... 0.7930290699005127 seconds
run: ('random_01_run1_all_corrected_noise_rn5ry0.pt', 01_generated_0) ... 0.7605545520782471 seconds
run: ('random_01_run1_all_corrected_noise_rn5ry0.pt', 01_generated_1) ... 0.7493987083435059 seconds
run: ('random_01_run1_scope_corrected_noise_rn5ry0.pt', 01_generated_0) ... 0.8135294914245605 seconds
run: ('random_01_run1_scope_corrected_noise_rn5ry0.pt', 01_generated_1) ... 0.7734940052032471 seconds
run: ('random_01_run1_noneut_corrected_rn5ry0.pt', 01_generated_0) ... 0.7779641151428223 seconds
run: ('random_01_run1_noneut_corrected_rn5ry0.pt', 01_generated_1) ... 0.7690799236297607 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn5ry0.pt', 01_generated_0) ... 0.8127131462097168 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn5ry0.pt', 01_generated_1) ... 0.7881405353546143 seconds
run: ('random_01_run1_allnoise_correctedscore_rn5ry0.pt', 01_generated_0) ... 0.8090994358062744 seconds
run: ('random_01_run1_allnoise_correctedscore_rn5ry0.pt', 01_generated_1) ... 0.7811079025268555 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn5ry0.pt', 01_generated_0) ... 0.8154122829437256 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn5ry0.pt', 01_generated_1) ... 0.7895808219909668 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn5ry0.pt', 01_generated_0) ... 0.7696430683135986 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn5ry0.pt', 01_generated_1) ... 0.7702295780181885 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn5ry0.pt', 01_generated_0) ... 0.814948558807373 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn5ry0.pt', 01_generated_1) ... 0.781724214553833 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn5ry0.pt', 01_generated_0) ... 0.8129715919494629 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn5ry0.pt', 01_generated_1) ... 0.7851817607879639 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn5ry0.pt', 01_generated_0) ... 0.8112649917602539 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn5ry0.pt', 01_generated_1) ... 0.7729573249816895 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn5ry0.pt', 01_generated_0) ... 0.7883114814758301 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn5ry0.pt', 01_generated_1) ... 0.7861919403076172 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn5ry0.pt', 01_generated_0) ... 0.8524191379547119 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn5ry0.pt', 01_generated_1) ... 0.8037223815917969 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn5ry0.pt', 01_generated_0) ... 0.7701647281646729 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn5ry0.pt', 01_generated_1) ... 0.7749607563018799 seconds
run: ('random_01_run1_all_rn6ry0.pt', 01_generated_0) ... 0.7652673721313477 seconds
run: ('random_01_run1_all_rn6ry0.pt', 01_generated_1) ... 0.7312507629394531 seconds
run: ('random_01_run1_noneut_rn6ry0.pt', 01_generated_0) ... 0.7573401927947998 seconds
run: ('random_01_run1_noneut_rn6ry0.pt', 01_generated_1) ... 0.7939298152923584 seconds
run: ('random_01_run1_high_rn6ry0.pt', 01_generated_0) ... 0.7513625621795654 seconds
run: ('random_01_run1_high_rn6ry0.pt', 01_generated_1) ... 0.7557268142700195 seconds
run: ('random_01_run1_all_corrected_rn6ry0.pt', 01_generated_0) ... 0.7546114921569824 seconds
run: ('random_01_run1_all_corrected_rn6ry0.pt', 01_generated_1) ... 0.7843255996704102 seconds
run: ('random_01_run1_all_corrected_noise_rn6ry0.pt', 01_generated_0) ... 0.753425121307373 seconds
run: ('random_01_run1_all_corrected_noise_rn6ry0.pt', 01_generated_1) ... 0.7277154922485352 seconds
run: ('random_01_run1_scope_corrected_noise_rn6ry0.pt', 01_generated_0) ... 0.7514998912811279 seconds
run: ('random_01_run1_scope_corrected_noise_rn6ry0.pt', 01_generated_1) ... 0.7533724308013916 seconds
run: ('random_01_run1_noneut_corrected_rn6ry0.pt', 01_generated_0) ... 0.7719883918762207 seconds
run: ('random_01_run1_noneut_corrected_rn6ry0.pt', 01_generated_1) ... 0.7520318031311035 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn6ry0.pt', 01_generated_0) ... 0.7938268184661865 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn6ry0.pt', 01_generated_1) ... 0.7360999584197998 seconds
run: ('random_01_run1_allnoise_correctedscore_rn6ry0.pt', 01_generated_0) ... 0.7569401264190674 seconds
run: ('random_01_run1_allnoise_correctedscore_rn6ry0.pt', 01_generated_1) ... 0.7233860492706299 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn6ry0.pt', 01_generated_0) ... 0.7502574920654297 seconds
 
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn6ry0.pt', 01_generated_1) ... 0.7246582508087158 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn6ry0.pt', 01_generated_0) ... 0.7902188301086426 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn6ry0.pt', 01_generated_1) ... 0.7236385345458984 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn6ry0.pt', 01_generated_0) ... 0.7912733554840088 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn6ry0.pt', 01_generated_1) ... 0.7643671035766602 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn6ry0.pt', 01_generated_0) ... 0.8239908218383789 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn6ry0.pt', 01_generated_1) ... 0.7449631690979004 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn6ry0.pt', 01_generated_0) ... 0.7789182662963867 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn6ry0.pt', 01_generated_1) ... 0.7656729221343994 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn6ry0.pt', 01_generated_0) ... 0.8134493827819824 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn6ry0.pt', 01_generated_1) ... 0.7495450973510742 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn6ry0.pt', 01_generated_0) ... 0.8074796199798584 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn6ry0.pt', 01_generated_1) ... 0.7580208778381348 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn6ry0.pt', 01_generated_0) ... 0.8038129806518555 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn6ry0.pt', 01_generated_1) ... 0.7241384983062744 seconds
run: ('random_01_run1_all_rn7ry0.pt', 01_generated_0) ... 0.753338098526001 seconds
run: ('random_01_run1_all_rn7ry0.pt', 01_generated_1) ... 0.7256934642791748 seconds
run: ('random_01_run1_noneut_rn7ry0.pt', 01_generated_0) ... 0.7960789203643799 seconds
run: ('random_01_run1_noneut_rn7ry0.pt', 01_generated_1) ... 0.725600004196167 seconds
run: ('random_01_run1_high_rn7ry0.pt', 01_generated_0) ... 0.7538223266601562 seconds
run: ('random_01_run1_high_rn7ry0.pt', 01_generated_1) ... 0.7262768745422363 seconds
run: ('random_01_run1_all_corrected_rn7ry0.pt', 01_generated_0) ... 0.7642257213592529 seconds
run: ('random_01_run1_all_corrected_rn7ry0.pt', 01_generated_1) ... 0.7224369049072266 seconds
run: ('random_01_run1_all_corrected_noise_rn7ry0.pt', 01_generated_0) ... 0.7482075691223145 seconds
run: ('random_01_run1_all_corrected_noise_rn7ry0.pt', 01_generated_1) ... 0.7210140228271484 seconds
run: ('random_01_run1_scope_corrected_noise_rn7ry0.pt', 01_generated_0) ... 0.7659103870391846 seconds
run: ('random_01_run1_scope_corrected_noise_rn7ry0.pt', 01_generated_1) ... 0.7236132621765137 seconds
run: ('random_01_run1_noneut_corrected_rn7ry0.pt', 01_generated_0) ... 0.7787466049194336 seconds
run: ('random_01_run1_noneut_corrected_rn7ry0.pt', 01_generated_1) ... 0.7912085056304932 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn7ry0.pt', 01_generated_0) ... 0.7605891227722168 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn7ry0.pt', 01_generated_1) ... 0.7304017543792725 seconds
run: ('random_01_run1_allnoise_correctedscore_rn7ry0.pt', 01_generated_0) ... 0.7506458759307861 seconds
run: ('random_01_run1_allnoise_correctedscore_rn7ry0.pt', 01_generated_1) ... 0.7258670330047607 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn7ry0.pt', 01_generated_0) ... 0.7548160552978516 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn7ry0.pt', 01_generated_1) ... 0.7656910419464111 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn7ry0.pt', 01_generated_0) ... 0.751469612121582 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn7ry0.pt', 01_generated_1) ... 0.7233421802520752 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn7ry0.pt', 01_generated_0) ... 0.7656795978546143 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn7ry0.pt', 01_generated_1) ... 0.7848131656646729 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn7ry0.pt', 01_generated_0) ... 0.7897906303405762 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn7ry0.pt', 01_generated_1) ... 0.7379364967346191 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn7ry0.pt', 01_generated_0) ... 0.7927961349487305 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn7ry0.pt', 01_generated_1) ... 0.7816534042358398 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn7ry0.pt', 01_generated_0) ... 0.7563142776489258 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn7ry0.pt', 01_generated_1) ... 0.7225344181060791 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn7ry0.pt', 01_generated_0) ... 0.7529323101043701 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn7ry0.pt', 01_generated_1) ... 0.7419722080230713 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn7ry0.pt', 01_generated_0) ... 0.7957742214202881 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn7ry0.pt', 01_generated_1) ... 0.7268612384796143 seconds
run: ('random_01_run1_all_rn8ry0.pt', 01_generated_0) ... 0.7521834373474121 seconds
run: ('random_01_run1_all_rn8ry0.pt', 01_generated_1) ... 0.7610480785369873 seconds
run: ('random_01_run1_noneut_rn8ry0.pt', 01_generated_0) ... 0.7592580318450928 seconds
run: ('random_01_run1_noneut_rn8ry0.pt', 01_generated_1) ... 0.7240767478942871 seconds
run: ('random_01_run1_high_rn8ry0.pt', 01_generated_0) ... 0.7542121410369873 seconds
run: ('random_01_run1_high_rn8ry0.pt', 01_generated_1) ... 0.7459311485290527 seconds
run: ('random_01_run1_all_corrected_rn8ry0.pt', 01_generated_0) ... 0.7751269340515137 seconds
run: ('random_01_run1_all_corrected_rn8ry0.pt', 01_generated_1) ... 0.7238273620605469 seconds
run: ('random_01_run1_all_corrected_noise_rn8ry0.pt', 01_generated_0) ... 0.7565290927886963 seconds
run: ('random_01_run1_all_corrected_noise_rn8ry0.pt', 01_generated_1) ... 0.7399177551269531 seconds
run: ('random_01_run1_scope_corrected_noise_rn8ry0.pt', 01_generated_0) ... 0.7918801307678223 seconds
run: ('random_01_run1_scope_corrected_noise_rn8ry0.pt', 01_generated_1) ... 0.7243120670318604 seconds
run: ('random_01_run1_noneut_corrected_rn8ry0.pt', 01_generated_0) ... 0.7518818378448486 seconds
run: ('random_01_run1_noneut_corrected_rn8ry0.pt', 01_generated_1) ... 0.72422194480896 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn8ry0.pt', 01_generated_0) ... 0.8041338920593262 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn8ry0.pt', 01_generated_1) ... 0.7337734699249268 seconds
run: ('random_01_run1_allnoise_correctedscore_rn8ry0.pt', 01_generated_0) ... 0.755748987197876 seconds
run: ('random_01_run1_allnoise_correctedscore_rn8ry0.pt', 01_generated_1) ... 0.7497692108154297 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn8ry0.pt', 01_generated_0) ... 0.7966079711914062 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn8ry0.pt', 01_generated_1) ... 0.7259902954101562 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn8ry0.pt', 01_generated_0) ... 0.7603096961975098 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn8ry0.pt', 01_generated_1) ... 0.7311968803405762 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn8ry0.pt', 01_generated_0) ... 0.8043205738067627 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn8ry0.pt', 01_generated_1) ... 0.7284860610961914 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn8ry0.pt', 01_generated_0) ... 0.7527167797088623 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn8ry0.pt', 01_generated_1) ... 0.7266857624053955 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn8ry0.pt', 01_generated_0) ... 0.7870209217071533 seconds
 
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn8ry0.pt', 01_generated_1) ... 0.7302298545837402 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn8ry0.pt', 01_generated_0) ... 0.7540531158447266 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn8ry0.pt', 01_generated_1) ... 0.7243797779083252 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn8ry0.pt', 01_generated_0) ... 0.7680184841156006 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn8ry0.pt', 01_generated_1) ... 0.7375857830047607 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn8ry0.pt', 01_generated_0) ... 0.7470803260803223 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn8ry0.pt', 01_generated_1) ... 0.7253513336181641 seconds
run: ('random_01_run1_all_rn9ry0.pt', 01_generated_0) ... 0.7560834884643555 seconds
run: ('random_01_run1_all_rn9ry0.pt', 01_generated_1) ... 0.767967700958252 seconds
run: ('random_01_run1_noneut_rn9ry0.pt', 01_generated_0) ... 0.7544729709625244 seconds
run: ('random_01_run1_noneut_rn9ry0.pt', 01_generated_1) ... 0.7218246459960938 seconds
run: ('random_01_run1_high_rn9ry0.pt', 01_generated_0) ... 0.7526724338531494 seconds
run: ('random_01_run1_high_rn9ry0.pt', 01_generated_1) ... 0.7375478744506836 seconds
run: ('random_01_run1_all_corrected_rn9ry0.pt', 01_generated_0) ... 0.7505922317504883 seconds
run: ('random_01_run1_all_corrected_rn9ry0.pt', 01_generated_1) ... 0.7195403575897217 seconds
run: ('random_01_run1_all_corrected_noise_rn9ry0.pt', 01_generated_0) ... 0.7532250881195068 seconds
run: ('random_01_run1_all_corrected_noise_rn9ry0.pt', 01_generated_1) ... 0.7672927379608154 seconds
run: ('random_01_run1_scope_corrected_noise_rn9ry0.pt', 01_generated_0) ... 0.7501945495605469 seconds
run: ('random_01_run1_scope_corrected_noise_rn9ry0.pt', 01_generated_1) ... 0.7194745540618896 seconds
run: ('random_01_run1_noneut_corrected_rn9ry0.pt', 01_generated_0) ... 0.7514503002166748 seconds
run: ('random_01_run1_noneut_corrected_rn9ry0.pt', 01_generated_1) ... 0.7522056102752686 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn9ry0.pt', 01_generated_0) ... 0.7568910121917725 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn9ry0.pt', 01_generated_1) ... 0.7291314601898193 seconds
run: ('random_01_run1_allnoise_correctedscore_rn9ry0.pt', 01_generated_0) ... 0.7542650699615479 seconds
run: ('random_01_run1_allnoise_correctedscore_rn9ry0.pt', 01_generated_1) ... 0.7447538375854492 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn9ry0.pt', 01_generated_0) ... 0.7731060981750488 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn9ry0.pt', 01_generated_1) ... 0.7330734729766846 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn9ry0.pt', 01_generated_0) ... 0.7541966438293457 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn9ry0.pt', 01_generated_1) ... 0.7322402000427246 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn9ry0.pt', 01_generated_0) ... 0.7888772487640381 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn9ry0.pt', 01_generated_1) ... 0.7210211753845215 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn9ry0.pt', 01_generated_0) ... 0.7540946006774902 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn9ry0.pt', 01_generated_1) ... 0.7261619567871094 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn9ry0.pt', 01_generated_0) ... 0.8265552520751953 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn9ry0.pt', 01_generated_1) ... 0.7251391410827637 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn9ry0.pt', 01_generated_0) ... 0.7898454666137695 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn9ry0.pt', 01_generated_1) ... 0.7711637020111084 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn9ry0.pt', 01_generated_0) ... 0.8068084716796875 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn9ry0.pt', 01_generated_1) ... 0.7555673122406006 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn9ry0.pt', 01_generated_0) ... 0.7794592380523682 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn9ry0.pt', 01_generated_1) ... 0.7421045303344727 seconds
run: ('random_01_run1_all_rn0ry0.pt', 01_generated_0) ... 0.7931544780731201 seconds
run: ('random_01_run1_all_rn0ry0.pt', 01_generated_1) ... 0.749375581741333 seconds
run: ('random_01_run1_noneut_rn0ry0.pt', 01_generated_0) ... 0.7810013294219971 seconds
run: ('random_01_run1_noneut_rn0ry0.pt', 01_generated_1) ... 0.7668750286102295 seconds
run: ('random_01_run1_high_rn0ry0.pt', 01_generated_0) ... 0.761420726776123 seconds
run: ('random_01_run1_high_rn0ry0.pt', 01_generated_1) ... 0.7466740608215332 seconds
run: ('random_01_run1_all_corrected_rn0ry0.pt', 01_generated_0) ... 0.8020205497741699 seconds
run: ('random_01_run1_all_corrected_rn0ry0.pt', 01_generated_1) ... 0.7746319770812988 seconds
run: ('random_01_run1_all_corrected_noise_rn0ry0.pt', 01_generated_0) ... 0.8085269927978516 seconds
run: ('random_01_run1_all_corrected_noise_rn0ry0.pt', 01_generated_1) ... 0.7538270950317383 seconds
run: ('random_01_run1_scope_corrected_noise_rn0ry0.pt', 01_generated_0) ... 0.7816188335418701 seconds
run: ('random_01_run1_scope_corrected_noise_rn0ry0.pt', 01_generated_1) ... 0.7484931945800781 seconds
run: ('random_01_run1_noneut_corrected_rn0ry0.pt', 01_generated_0) ... 0.7871768474578857 seconds
run: ('random_01_run1_noneut_corrected_rn0ry0.pt', 01_generated_1) ... 0.7272014617919922 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn0ry0.pt', 01_generated_0) ... 0.7507843971252441 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn0ry0.pt', 01_generated_1) ... 0.72422194480896 seconds
run: ('random_01_run1_allnoise_correctedscore_rn0ry0.pt', 01_generated_0) ... 0.7504775524139404 seconds
run: ('random_01_run1_allnoise_correctedscore_rn0ry0.pt', 01_generated_1) ... 0.7292976379394531 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn0ry0.pt', 01_generated_0) ... 0.7557790279388428 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn0ry0.pt', 01_generated_1) ... 0.7747366428375244 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn0ry0.pt', 01_generated_0) ... 0.7756223678588867 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn0ry0.pt', 01_generated_1) ... 0.7245681285858154 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn0ry0.pt', 01_generated_0) ... 0.7614321708679199 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn0ry0.pt', 01_generated_1) ... 0.7725498676300049 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn0ry0.pt', 01_generated_0) ... 0.7658045291900635 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn0ry0.pt', 01_generated_1) ... 0.7402698993682861 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn0ry0.pt', 01_generated_0) ... 0.8209044933319092 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn0ry0.pt', 01_generated_1) ... 0.7414579391479492 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry0.pt', 01_generated_0) ... 0.799858808517456 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry0.pt', 01_generated_1) ... 0.7381839752197266 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry0.pt', 01_generated_0) ... 0.7638349533081055 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry0.pt', 01_generated_1) ... 0.7589070796966553 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry0.pt', 01_generated_0) ... 0.7698507308959961 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry0.pt', 01_generated_1) ... 0.757737398147583 seconds
run: ('random_01_run1_all_rn0ry2.pt', 01_generated_0) ... 0.7720880508422852 seconds
 
run: ('random_01_run1_all_rn0ry2.pt', 01_generated_1) ... 0.7248127460479736 seconds
run: ('random_01_run1_noneut_rn0ry2.pt', 01_generated_0) ... 0.7541711330413818 seconds
run: ('random_01_run1_noneut_rn0ry2.pt', 01_generated_1) ... 0.7340245246887207 seconds
run: ('random_01_run1_high_rn0ry2.pt', 01_generated_0) ... 0.7589895725250244 seconds
run: ('random_01_run1_high_rn0ry2.pt', 01_generated_1) ... 0.7502892017364502 seconds
run: ('random_01_run1_all_corrected_rn0ry2.pt', 01_generated_0) ... 0.7699298858642578 seconds
run: ('random_01_run1_all_corrected_rn0ry2.pt', 01_generated_1) ... 0.7430582046508789 seconds
run: ('random_01_run1_all_corrected_noise_rn0ry2.pt', 01_generated_0) ... 0.7971148490905762 seconds
run: ('random_01_run1_all_corrected_noise_rn0ry2.pt', 01_generated_1) ... 0.7411129474639893 seconds
run: ('random_01_run1_scope_corrected_noise_rn0ry2.pt', 01_generated_0) ... 0.7744584083557129 seconds
run: ('random_01_run1_scope_corrected_noise_rn0ry2.pt', 01_generated_1) ... 0.7506372928619385 seconds
run: ('random_01_run1_noneut_corrected_rn0ry2.pt', 01_generated_0) ... 0.7801246643066406 seconds
run: ('random_01_run1_noneut_corrected_rn0ry2.pt', 01_generated_1) ... 0.7325921058654785 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn0ry2.pt', 01_generated_0) ... 0.7586774826049805 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn0ry2.pt', 01_generated_1) ... 0.7547626495361328 seconds
run: ('random_01_run1_allnoise_correctedscore_rn0ry2.pt', 01_generated_0) ... 0.8117153644561768 seconds
run: ('random_01_run1_allnoise_correctedscore_rn0ry2.pt', 01_generated_1) ... 0.7356126308441162 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn0ry2.pt', 01_generated_0) ... 0.7568249702453613 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn0ry2.pt', 01_generated_1) ... 0.7535953521728516 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn0ry2.pt', 01_generated_0) ... 0.7817118167877197 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn0ry2.pt', 01_generated_1) ... 0.7345309257507324 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn0ry2.pt', 01_generated_0) ... 0.750739336013794 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn0ry2.pt', 01_generated_1) ... 0.7531485557556152 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn0ry2.pt', 01_generated_0) ... 0.7823030948638916 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn0ry2.pt', 01_generated_1) ... 0.7481110095977783 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn0ry2.pt', 01_generated_0) ... 0.7780904769897461 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn0ry2.pt', 01_generated_1) ... 0.7447540760040283 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry2.pt', 01_generated_0) ... 0.7580869197845459 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry2.pt', 01_generated_1) ... 0.74159836769104 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry2.pt', 01_generated_0) ... 0.7740566730499268 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry2.pt', 01_generated_1) ... 0.7349162101745605 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry2.pt', 01_generated_0) ... 0.7576632499694824 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry2.pt', 01_generated_1) ... 0.7281792163848877 seconds
run: ('random_01_run1_all_rn0ry3.pt', 01_generated_0) ... 0.7622473239898682 seconds
run: ('random_01_run1_all_rn0ry3.pt', 01_generated_1) ... 0.7284712791442871 seconds
run: ('random_01_run1_noneut_rn0ry3.pt', 01_generated_0) ... 0.7544634342193604 seconds
run: ('random_01_run1_noneut_rn0ry3.pt', 01_generated_1) ... 0.7440564632415771 seconds
run: ('random_01_run1_high_rn0ry3.pt', 01_generated_0) ... 0.7993321418762207 seconds
run: ('random_01_run1_high_rn0ry3.pt', 01_generated_1) ... 0.7295231819152832 seconds
run: ('random_01_run1_all_corrected_rn0ry3.pt', 01_generated_0) ... 0.7764503955841064 seconds
run: ('random_01_run1_all_corrected_rn0ry3.pt', 01_generated_1) ... 0.7351126670837402 seconds
run: ('random_01_run1_all_corrected_noise_rn0ry3.pt', 01_generated_0) ... 0.7847855091094971 seconds
run: ('random_01_run1_all_corrected_noise_rn0ry3.pt', 01_generated_1) ... 0.7405533790588379 seconds
run: ('random_01_run1_scope_corrected_noise_rn0ry3.pt', 01_generated_0) ... 0.7681460380554199 seconds
run: ('random_01_run1_scope_corrected_noise_rn0ry3.pt', 01_generated_1) ... 0.7301697731018066 seconds
run: ('random_01_run1_noneut_corrected_rn0ry3.pt', 01_generated_0) ... 0.7874753475189209 seconds
run: ('random_01_run1_noneut_corrected_rn0ry3.pt', 01_generated_1) ... 0.7408134937286377 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn0ry3.pt', 01_generated_0) ... 0.7938282489776611 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn0ry3.pt', 01_generated_1) ... 0.752943754196167 seconds
run: ('random_01_run1_allnoise_correctedscore_rn0ry3.pt', 01_generated_0) ... 0.7818329334259033 seconds
run: ('random_01_run1_allnoise_correctedscore_rn0ry3.pt', 01_generated_1) ... 0.7411091327667236 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn0ry3.pt', 01_generated_0) ... 0.7711944580078125 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn0ry3.pt', 01_generated_1) ... 0.7364246845245361 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn0ry3.pt', 01_generated_0) ... 0.7649140357971191 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn0ry3.pt', 01_generated_1) ... 0.7401509284973145 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn0ry3.pt', 01_generated_0) ... 0.7640080451965332 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn0ry3.pt', 01_generated_1) ... 0.736840009689331 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn0ry3.pt', 01_generated_0) ... 0.7616758346557617 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn0ry3.pt', 01_generated_1) ... 0.7367570400238037 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn0ry3.pt', 01_generated_0) ... 0.7623269557952881 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn0ry3.pt', 01_generated_1) ... 0.7341678142547607 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry3.pt', 01_generated_0) ... 0.7584352493286133 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry3.pt', 01_generated_1) ... 0.7435369491577148 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry3.pt', 01_generated_0) ... 0.7560133934020996 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry3.pt', 01_generated_1) ... 0.7342803478240967 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry3.pt', 01_generated_0) ... 0.7616426944732666 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry3.pt', 01_generated_1) ... 0.7371120452880859 seconds
run: ('random_01_run1_all_rn0ry4.pt', 01_generated_0) ... 0.7661745548248291 seconds
run: ('random_01_run1_all_rn0ry4.pt', 01_generated_1) ... 0.7287240028381348 seconds
run: ('random_01_run1_noneut_rn0ry4.pt', 01_generated_0) ... 0.7579519748687744 seconds
run: ('random_01_run1_noneut_rn0ry4.pt', 01_generated_1) ... 0.7341234683990479 seconds
run: ('random_01_run1_high_rn0ry4.pt', 01_generated_0) ... 0.759737491607666 seconds
run: ('random_01_run1_high_rn0ry4.pt', 01_generated_1) ... 0.733389139175415 seconds
run: ('random_01_run1_all_corrected_rn0ry4.pt', 01_generated_0) ... 0.7559244632720947 seconds
run: ('random_01_run1_all_corrected_rn0ry4.pt', 01_generated_1) ... 0.7311429977416992 seconds
run: ('random_01_run1_all_corrected_noise_rn0ry4.pt', 01_generated_0) ... 0.7553586959838867 seconds
run: ('random_01_run1_all_corrected_noise_rn0ry4.pt', 01_generated_1) ... 0.7417051792144775 seconds
run: ('random_01_run1_scope_corrected_noise_rn0ry4.pt', 01_generated_0) ... 0.7928597927093506 seconds
 
run: ('random_01_run1_scope_corrected_noise_rn0ry4.pt', 01_generated_1) ... 0.7308485507965088 seconds
run: ('random_01_run1_noneut_corrected_rn0ry4.pt', 01_generated_0) ... 0.7528059482574463 seconds
run: ('random_01_run1_noneut_corrected_rn0ry4.pt', 01_generated_1) ... 0.7271997928619385 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn0ry4.pt', 01_generated_0) ... 0.756000280380249 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn0ry4.pt', 01_generated_1) ... 0.7263185977935791 seconds
run: ('random_01_run1_allnoise_correctedscore_rn0ry4.pt', 01_generated_0) ... 0.7840168476104736 seconds
run: ('random_01_run1_allnoise_correctedscore_rn0ry4.pt', 01_generated_1) ... 0.733933687210083 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn0ry4.pt', 01_generated_0) ... 0.756371259689331 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn0ry4.pt', 01_generated_1) ... 0.7248344421386719 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn0ry4.pt', 01_generated_0) ... 0.7510194778442383 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn0ry4.pt', 01_generated_1) ... 0.7240080833435059 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn0ry4.pt', 01_generated_0) ... 0.748488187789917 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn0ry4.pt', 01_generated_1) ... 0.7299108505249023 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn0ry4.pt', 01_generated_0) ... 0.7557837963104248 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn0ry4.pt', 01_generated_1) ... 0.7270498275756836 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn0ry4.pt', 01_generated_0) ... 0.7859997749328613 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn0ry4.pt', 01_generated_1) ... 0.7267389297485352 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry4.pt', 01_generated_0) ... 0.7674970626831055 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry4.pt', 01_generated_1) ... 0.7261285781860352 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry4.pt', 01_generated_0) ... 0.7892324924468994 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry4.pt', 01_generated_1) ... 0.720534086227417 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry4.pt', 01_generated_0) ... 0.7500591278076172 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry4.pt', 01_generated_1) ... 0.7246818542480469 seconds
run: ('random_01_run1_all_rn0ry5.pt', 01_generated_0) ... 0.7533166408538818 seconds
run: ('random_01_run1_all_rn0ry5.pt', 01_generated_1) ... 0.7223837375640869 seconds
run: ('random_01_run1_noneut_rn0ry5.pt', 01_generated_0) ... 0.7539429664611816 seconds
run: ('random_01_run1_noneut_rn0ry5.pt', 01_generated_1) ... 0.7239990234375 seconds
run: ('random_01_run1_high_rn0ry5.pt', 01_generated_0) ... 0.7494180202484131 seconds
run: ('random_01_run1_high_rn0ry5.pt', 01_generated_1) ... 0.7243328094482422 seconds
run: ('random_01_run1_all_corrected_rn0ry5.pt', 01_generated_0) ... 0.7539427280426025 seconds
run: ('random_01_run1_all_corrected_rn0ry5.pt', 01_generated_1) ... 0.7242612838745117 seconds
run: ('random_01_run1_all_corrected_noise_rn0ry5.pt', 01_generated_0) ... 0.7529010772705078 seconds
run: ('random_01_run1_all_corrected_noise_rn0ry5.pt', 01_generated_1) ... 0.7240128517150879 seconds
run: ('random_01_run1_scope_corrected_noise_rn0ry5.pt', 01_generated_0) ... 0.756493091583252 seconds
run: ('random_01_run1_scope_corrected_noise_rn0ry5.pt', 01_generated_1) ... 0.728125810623169 seconds
run: ('random_01_run1_noneut_corrected_rn0ry5.pt', 01_generated_0) ... 0.7552022933959961 seconds
run: ('random_01_run1_noneut_corrected_rn0ry5.pt', 01_generated_1) ... 0.7296204566955566 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn0ry5.pt', 01_generated_0) ... 0.7609193325042725 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn0ry5.pt', 01_generated_1) ... 0.732264518737793 seconds
run: ('random_01_run1_allnoise_correctedscore_rn0ry5.pt', 01_generated_0) ... 0.7565100193023682 seconds
run: ('random_01_run1_allnoise_correctedscore_rn0ry5.pt', 01_generated_1) ... 0.7283315658569336 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn0ry5.pt', 01_generated_0) ... 0.7532527446746826 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn0ry5.pt', 01_generated_1) ... 0.7292211055755615 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn0ry5.pt', 01_generated_0) ... 0.7564864158630371 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn0ry5.pt', 01_generated_1) ... 0.7276005744934082 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn0ry5.pt', 01_generated_0) ... 0.7538971900939941 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn0ry5.pt', 01_generated_1) ... 0.7252380847930908 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn0ry5.pt', 01_generated_0) ... 0.7564146518707275 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn0ry5.pt', 01_generated_1) ... 0.736487865447998 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn0ry5.pt', 01_generated_0) ... 0.751784086227417 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn0ry5.pt', 01_generated_1) ... 0.7327902317047119 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry5.pt', 01_generated_0) ... 0.756572961807251 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry5.pt', 01_generated_1) ... 0.7279584407806396 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry5.pt', 01_generated_0) ... 0.7560429573059082 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry5.pt', 01_generated_1) ... 0.7347691059112549 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry5.pt', 01_generated_0) ... 0.7596995830535889 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry5.pt', 01_generated_1) ... 0.7253303527832031 seconds
run: ('random_01_run1_all_rn0ry6.pt', 01_generated_0) ... 0.7534193992614746 seconds
run: ('random_01_run1_all_rn0ry6.pt', 01_generated_1) ... 0.7271809577941895 seconds
run: ('random_01_run1_noneut_rn0ry6.pt', 01_generated_0) ... 0.7547793388366699 seconds
run: ('random_01_run1_noneut_rn0ry6.pt', 01_generated_1) ... 0.7273962497711182 seconds
run: ('random_01_run1_high_rn0ry6.pt', 01_generated_0) ... 0.7566056251525879 seconds
run: ('random_01_run1_high_rn0ry6.pt', 01_generated_1) ... 0.7289373874664307 seconds
run: ('random_01_run1_all_corrected_rn0ry6.pt', 01_generated_0) ... 0.753943920135498 seconds
run: ('random_01_run1_all_corrected_rn0ry6.pt', 01_generated_1) ... 0.7255895137786865 seconds
run: ('random_01_run1_all_corrected_noise_rn0ry6.pt', 01_generated_0) ... 0.7541353702545166 seconds
run: ('random_01_run1_all_corrected_noise_rn0ry6.pt', 01_generated_1) ... 0.7274472713470459 seconds
run: ('random_01_run1_scope_corrected_noise_rn0ry6.pt', 01_generated_0) ... 0.7508771419525146 seconds
run: ('random_01_run1_scope_corrected_noise_rn0ry6.pt', 01_generated_1) ... 0.7262105941772461 seconds
run: ('random_01_run1_noneut_corrected_rn0ry6.pt', 01_generated_0) ... 0.7519607543945312 seconds
run: ('random_01_run1_noneut_corrected_rn0ry6.pt', 01_generated_1) ... 0.7242169380187988 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn0ry6.pt', 01_generated_0) ... 0.754462480545044 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn0ry6.pt', 01_generated_1) ... 0.7292478084564209 seconds
run: ('random_01_run1_allnoise_correctedscore_rn0ry6.pt', 01_generated_0) ... 0.7564153671264648 seconds
run: ('random_01_run1_allnoise_correctedscore_rn0ry6.pt', 01_generated_1) ... 0.7243220806121826 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn0ry6.pt', 01_generated_0) ... 0.754986047744751 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn0ry6.pt', 01_generated_1) ... 0.7318720817565918 seconds
 
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn0ry6.pt', 01_generated_0) ... 0.7593591213226318 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn0ry6.pt', 01_generated_1) ... 0.7321429252624512 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn0ry6.pt', 01_generated_0) ... 0.7553620338439941 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn0ry6.pt', 01_generated_1) ... 0.7293384075164795 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn0ry6.pt', 01_generated_0) ... 0.75602126121521 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn0ry6.pt', 01_generated_1) ... 0.7275826930999756 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn0ry6.pt', 01_generated_0) ... 0.7559683322906494 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn0ry6.pt', 01_generated_1) ... 0.7261447906494141 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry6.pt', 01_generated_0) ... 0.7575318813323975 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry6.pt', 01_generated_1) ... 0.726238489151001 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry6.pt', 01_generated_0) ... 0.7658705711364746 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry6.pt', 01_generated_1) ... 0.7270655632019043 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry6.pt', 01_generated_0) ... 0.7544467449188232 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry6.pt', 01_generated_1) ... 0.7271747589111328 seconds
run: ('random_01_run1_all_rn0ry7.pt', 01_generated_0) ... 0.7600588798522949 seconds
run: ('random_01_run1_all_rn0ry7.pt', 01_generated_1) ... 0.7294116020202637 seconds
run: ('random_01_run1_noneut_rn0ry7.pt', 01_generated_0) ... 0.756502628326416 seconds
run: ('random_01_run1_noneut_rn0ry7.pt', 01_generated_1) ... 0.726351261138916 seconds
run: ('random_01_run1_high_rn0ry7.pt', 01_generated_0) ... 0.7559850215911865 seconds
run: ('random_01_run1_high_rn0ry7.pt', 01_generated_1) ... 0.7296712398529053 seconds
run: ('random_01_run1_all_corrected_rn0ry7.pt', 01_generated_0) ... 0.7613246440887451 seconds
run: ('random_01_run1_all_corrected_rn0ry7.pt', 01_generated_1) ... 0.7319555282592773 seconds
run: ('random_01_run1_all_corrected_noise_rn0ry7.pt', 01_generated_0) ... 0.7567348480224609 seconds
run: ('random_01_run1_all_corrected_noise_rn0ry7.pt', 01_generated_1) ... 0.7324156761169434 seconds
run: ('random_01_run1_scope_corrected_noise_rn0ry7.pt', 01_generated_0) ... 0.758925199508667 seconds
run: ('random_01_run1_scope_corrected_noise_rn0ry7.pt', 01_generated_1) ... 0.728135347366333 seconds
run: ('random_01_run1_noneut_corrected_rn0ry7.pt', 01_generated_0) ... 0.7592067718505859 seconds
run: ('random_01_run1_noneut_corrected_rn0ry7.pt', 01_generated_1) ... 0.7327067852020264 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn0ry7.pt', 01_generated_0) ... 0.7575953006744385 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn0ry7.pt', 01_generated_1) ... 0.724066972732544 seconds
run: ('random_01_run1_allnoise_correctedscore_rn0ry7.pt', 01_generated_0) ... 0.7546780109405518 seconds
run: ('random_01_run1_allnoise_correctedscore_rn0ry7.pt', 01_generated_1) ... 0.7291491031646729 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn0ry7.pt', 01_generated_0) ... 0.756633996963501 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn0ry7.pt', 01_generated_1) ... 0.7265863418579102 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn0ry7.pt', 01_generated_0) ... 0.7525761127471924 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn0ry7.pt', 01_generated_1) ... 0.7272839546203613 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn0ry7.pt', 01_generated_0) ... 0.75417160987854 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn0ry7.pt', 01_generated_1) ... 0.726233959197998 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn0ry7.pt', 01_generated_0) ... 0.750115156173706 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn0ry7.pt', 01_generated_1) ... 0.7237362861633301 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn0ry7.pt', 01_generated_0) ... 0.755089282989502 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn0ry7.pt', 01_generated_1) ... 0.7227497100830078 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry7.pt', 01_generated_0) ... 0.7542037963867188 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry7.pt', 01_generated_1) ... 0.7369389533996582 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry7.pt', 01_generated_0) ... 0.7538924217224121 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry7.pt', 01_generated_1) ... 0.7306926250457764 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry7.pt', 01_generated_0) ... 0.7554984092712402 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry7.pt', 01_generated_1) ... 0.7245688438415527 seconds
run: ('random_01_run1_all_rn0ry8.pt', 01_generated_0) ... 0.7552459239959717 seconds
run: ('random_01_run1_all_rn0ry8.pt', 01_generated_1) ... 0.7244765758514404 seconds
run: ('random_01_run1_noneut_rn0ry8.pt', 01_generated_0) ... 0.7546858787536621 seconds
run: ('random_01_run1_noneut_rn0ry8.pt', 01_generated_1) ... 0.724989652633667 seconds
run: ('random_01_run1_high_rn0ry8.pt', 01_generated_0) ... 0.7547485828399658 seconds
run: ('random_01_run1_high_rn0ry8.pt', 01_generated_1) ... 0.7289762496948242 seconds
run: ('random_01_run1_all_corrected_rn0ry8.pt', 01_generated_0) ... 0.7588112354278564 seconds
run: ('random_01_run1_all_corrected_rn0ry8.pt', 01_generated_1) ... 0.7321727275848389 seconds
run: ('random_01_run1_all_corrected_noise_rn0ry8.pt', 01_generated_0) ... 0.7570044994354248 seconds
run: ('random_01_run1_all_corrected_noise_rn0ry8.pt', 01_generated_1) ... 0.7262113094329834 seconds
run: ('random_01_run1_scope_corrected_noise_rn0ry8.pt', 01_generated_0) ... 0.7560422420501709 seconds
run: ('random_01_run1_scope_corrected_noise_rn0ry8.pt', 01_generated_1) ... 0.7259533405303955 seconds
run: ('random_01_run1_noneut_corrected_rn0ry8.pt', 01_generated_0) ... 0.7517969608306885 seconds
run: ('random_01_run1_noneut_corrected_rn0ry8.pt', 01_generated_1) ... 0.7279675006866455 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn0ry8.pt', 01_generated_0) ... 0.7553389072418213 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn0ry8.pt', 01_generated_1) ... 0.7243044376373291 seconds
run: ('random_01_run1_allnoise_correctedscore_rn0ry8.pt', 01_generated_0) ... 0.7541599273681641 seconds
run: ('random_01_run1_allnoise_correctedscore_rn0ry8.pt', 01_generated_1) ... 0.7278106212615967 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn0ry8.pt', 01_generated_0) ... 0.7496757507324219 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn0ry8.pt', 01_generated_1) ... 0.7258667945861816 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn0ry8.pt', 01_generated_0) ... 0.7542033195495605 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn0ry8.pt', 01_generated_1) ... 0.7233843803405762 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn0ry8.pt', 01_generated_0) ... 0.7569417953491211 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn0ry8.pt', 01_generated_1) ... 0.7273011207580566 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn0ry8.pt', 01_generated_0) ... 0.7612824440002441 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn0ry8.pt', 01_generated_1) ... 0.7315208911895752 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn0ry8.pt', 01_generated_0) ... 0.8197259902954102 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn0ry8.pt', 01_generated_1) ... 0.7639796733856201 seconds
 
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry8.pt', 01_generated_0) ... 0.7535443305969238 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry8.pt', 01_generated_1) ... 0.7246527671813965 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry8.pt', 01_generated_0) ... 0.7740280628204346 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry8.pt', 01_generated_1) ... 0.7435109615325928 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry8.pt', 01_generated_0) ... 0.7837469577789307 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry8.pt', 01_generated_1) ... 0.7285308837890625 seconds
run: ('random_01_run1_all_rn0ry9.pt', 01_generated_0) ... 0.7921938896179199 seconds
run: ('random_01_run1_all_rn0ry9.pt', 01_generated_1) ... 0.7515230178833008 seconds
run: ('random_01_run1_noneut_rn0ry9.pt', 01_generated_0) ... 0.8025062084197998 seconds
run: ('random_01_run1_noneut_rn0ry9.pt', 01_generated_1) ... 0.7435953617095947 seconds
run: ('random_01_run1_high_rn0ry9.pt', 01_generated_0) ... 0.7583441734313965 seconds
run: ('random_01_run1_high_rn0ry9.pt', 01_generated_1) ... 0.7263915538787842 seconds
run: ('random_01_run1_all_corrected_rn0ry9.pt', 01_generated_0) ... 0.7556836605072021 seconds
run: ('random_01_run1_all_corrected_rn0ry9.pt', 01_generated_1) ... 0.7479240894317627 seconds
run: ('random_01_run1_all_corrected_noise_rn0ry9.pt', 01_generated_0) ... 0.7595236301422119 seconds
run: ('random_01_run1_all_corrected_noise_rn0ry9.pt', 01_generated_1) ... 0.7346436977386475 seconds
run: ('random_01_run1_scope_corrected_noise_rn0ry9.pt', 01_generated_0) ... 0.7576563358306885 seconds
run: ('random_01_run1_scope_corrected_noise_rn0ry9.pt', 01_generated_1) ... 0.7307660579681396 seconds
run: ('random_01_run1_noneut_corrected_rn0ry9.pt', 01_generated_0) ... 0.7544121742248535 seconds
run: ('random_01_run1_noneut_corrected_rn0ry9.pt', 01_generated_1) ... 0.7278764247894287 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn0ry9.pt', 01_generated_0) ... 0.755192756652832 seconds
run: ('random_01_run1_allnoise_correctedhwgt_rn0ry9.pt', 01_generated_1) ... 0.7274973392486572 seconds
run: ('random_01_run1_allnoise_correctedscore_rn0ry9.pt', 01_generated_0) ... 0.7551116943359375 seconds
run: ('random_01_run1_allnoise_correctedscore_rn0ry9.pt', 01_generated_1) ... 0.7750239372253418 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn0ry9.pt', 01_generated_0) ... 0.7833237648010254 seconds
run: ('random_01_run1_allnoise_correctbydeepconvfilter_rn0ry9.pt', 01_generated_1) ... 0.727121114730835 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn0ry9.pt', 01_generated_0) ... 0.7878110408782959 seconds
run: ('random_01_run1_allnoise_correctbyfcndaefilter_rn0ry9.pt', 01_generated_1) ... 0.7507467269897461 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn0ry9.pt', 01_generated_0) ... 0.7537708282470703 seconds
run: ('random_01_run1_allnoise_correctbyconvlstmfilter_rn0ry9.pt', 01_generated_1) ... 0.745764970779419 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn0ry9.pt', 01_generated_0) ... 0.7598302364349365 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm2filter_rn0ry9.pt', 01_generated_1) ... 0.7436184883117676 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn0ry9.pt', 01_generated_0) ... 0.7500631809234619 seconds
run: ('random_01_run1_allnoise_correctbyconvlstm3filter_rn0ry9.pt', 01_generated_1) ... 0.7381124496459961 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry9.pt', 01_generated_0) ... 0.7494466304779053 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry9.pt', 01_generated_1) ... 0.7250678539276123 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry9.pt', 01_generated_0) ... 0.7515983581542969 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry9.pt', 01_generated_1) ... 0.72646164894104 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry9.pt', 01_generated_0) ... 0.7531909942626953 seconds
run: ('random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry9.pt', 01_generated_1) ... 0.738947868347168 seconds
saved predictions
 
%% Cell type:code id:94cc572d tags:
 
``` python
evaluation_manager.model_evaluation.models.keys()
```
 
%% Output
 
dict_keys(['HandWashingDeepConvLSTMA_trunc_01.pt', 'synthetic_inc_01.pt', '01_run1_all.pt', '01_run1_scope.pt', '01_run1_noneut.pt', '01_run1_high.pt', '01_run1_inchc.pt', '01_run1_all_corrected.pt', '01_run1_all_corrected_noise.pt', '01_run1_scope_corrected_noise.pt', '01_run1_scope_corrected.pt', '01_run1_noneut_corrected.pt', '01_run1_correctedhwgt.pt', '01_run1_noneut_correctedhwgt.pt', '01_run1_inchc_correctedhwgt.pt', '01_run1_synthetic.pt', '01_run1_allnoise_correctedhwgt.pt', '01_run1_allnoise_correctedhwgt_scope.pt', '01_run1_allnoise_corrected.pt', '01_run1_allnoise_correctedscore.pt', '01_run1_allnoise_correctedscore_flatten.pt', '01_run1_allnoise_correctbydeepconvfilter.pt', '01_run1_allnoise_correctbyfcndaefilter.pt', '01_run1_allnoise_correctbyfcndaefiltertest.pt', '01_run1_allnoise_correctbyconvlstmfilter.pt', '01_run1_allnoise_correctbyconvlstm2filter.pt', '01_run1_allnoise_correctbyconvlstm3filter.pt', '01_run1_alldeepconv_correctbyconvlstm3filter.pt', '01_run1_alldeepconv_correctbyconvlstm3filter2.pt', '01_run1_alldeepconv_correctbyconvlstm3filter3.pt', '01_run1_alldeepconv_correctbyconvlstm3filter4.pt', '01_run1_alldeepconv_correctbyconvlstm3filter5.pt', '01_run1_alldeepconv_correctbyconvlstm3filter6.pt', '01_run1_alldeepconv_correctbyconvlstm2filter6.pt', 'HandWashingDeepConvLSTMA_trunc_02.pt', 'synthetic_inc_02.pt', '02_run1_all.pt', '02_run1_scope.pt', '02_run1_noneut.pt', '02_run1_high.pt', '02_run1_inchc.pt', '02_run1_all_corrected.pt', '02_run1_all_corrected_noise.pt', '02_run1_scope_corrected_noise.pt', '02_run1_scope_corrected.pt', '02_run1_noneut_corrected.pt', '02_run1_correctedhwgt.pt', '02_run1_noneut_correctedhwgt.pt', '02_run1_inchc_correctedhwgt.pt', '02_run1_synthetic.pt', '02_run1_allnoise_correctedhwgt.pt', '02_run1_allnoise_correctedhwgt_scope.pt', '02_run1_allnoise_corrected.pt', '02_run1_allnoise_correctedscore.pt', '02_run1_allnoise_correctedscore_flatten.pt', '02_run1_allnoise_correctbydeepconvfilter.pt', '02_run1_allnoise_correctbyfcndaefilter.pt', '02_run1_allnoise_correctbyfcndaefiltertest.pt', '02_run1_allnoise_correctbyconvlstmfilter.pt', '02_run1_allnoise_correctbyconvlstm2filter.pt', '02_run1_allnoise_correctbyconvlstm3filter.pt', '02_run1_alldeepconv_correctbyconvlstm3filter.pt', '02_run1_alldeepconv_correctbyconvlstm3filter2.pt', '02_run1_alldeepconv_correctbyconvlstm3filter3.pt', '02_run1_alldeepconv_correctbyconvlstm3filter4.pt', '02_run1_alldeepconv_correctbyconvlstm3filter5.pt', '02_run1_alldeepconv_correctbyconvlstm3filter6.pt', '02_run1_alldeepconv_correctbyconvlstm2filter6.pt', 'HandWashingDeepConvLSTMA_trunc_10.pt', 'synthetic_inc_10.pt', '10_run1_all.pt', '10_run1_scope.pt', '10_run1_noneut.pt', '10_run1_high.pt', '10_run1_inchc.pt', '10_run1_all_corrected.pt', '10_run1_all_corrected_noise.pt', '10_run1_scope_corrected_noise.pt', '10_run1_scope_corrected.pt', '10_run1_noneut_corrected.pt', '10_run1_correctedhwgt.pt', '10_run1_noneut_correctedhwgt.pt', '10_run1_inchc_correctedhwgt.pt', '10_run1_synthetic.pt', '10_run1_allnoise_correctedhwgt.pt', '10_run1_allnoise_correctedhwgt_scope.pt', '10_run1_allnoise_corrected.pt', '10_run1_allnoise_correctedscore.pt', '10_run1_allnoise_correctedscore_flatten.pt', '10_run1_allnoise_correctbydeepconvfilter.pt', '10_run1_allnoise_correctbyfcndaefilter.pt', '10_run1_allnoise_correctbyfcndaefiltertest.pt', '10_run1_allnoise_correctbyconvlstmfilter.pt', '10_run1_allnoise_correctbyconvlstm2filter.pt', '10_run1_allnoise_correctbyconvlstm3filter.pt', '10_run1_alldeepconv_correctbyconvlstm3filter.pt', '10_run1_alldeepconv_correctbyconvlstm3filter2.pt', '10_run1_alldeepconv_correctbyconvlstm3filter3.pt', '10_run1_alldeepconv_correctbyconvlstm3filter4.pt', '10_run1_alldeepconv_correctbyconvlstm3filter5.pt', '10_run1_alldeepconv_correctbyconvlstm3filter6.pt', '10_run1_alldeepconv_correctbyconvlstm2filter6.pt', 'random_01_run1_all_rn2ry0.pt', 'random_01_run1_noneut_rn2ry0.pt', 'random_01_run1_high_rn2ry0.pt', 'random_01_run1_all_corrected_rn2ry0.pt', 'random_01_run1_all_corrected_noise_rn2ry0.pt', 'random_01_run1_scope_corrected_noise_rn2ry0.pt', 'random_01_run1_noneut_corrected_rn2ry0.pt', 'random_01_run1_allnoise_correctedhwgt_rn2ry0.pt', 'random_01_run1_allnoise_correctedscore_rn2ry0.pt', 'random_01_run1_allnoise_correctbydeepconvfilter_rn2ry0.pt', 'random_01_run1_allnoise_correctbyfcndaefilter_rn2ry0.pt', 'random_01_run1_allnoise_correctbyconvlstmfilter_rn2ry0.pt', 'random_01_run1_allnoise_correctbyconvlstm2filter_rn2ry0.pt', 'random_01_run1_allnoise_correctbyconvlstm3filter_rn2ry0.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter_rn2ry0.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn2ry0.pt', 'random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn2ry0.pt', 'random_01_run1_all_rn3ry0.pt', 'random_01_run1_noneut_rn3ry0.pt', 'random_01_run1_high_rn3ry0.pt', 'random_01_run1_all_corrected_rn3ry0.pt', 'random_01_run1_all_corrected_noise_rn3ry0.pt', 'random_01_run1_scope_corrected_noise_rn3ry0.pt', 'random_01_run1_noneut_corrected_rn3ry0.pt', 'random_01_run1_allnoise_correctedhwgt_rn3ry0.pt', 'random_01_run1_allnoise_correctedscore_rn3ry0.pt', 'random_01_run1_allnoise_correctbydeepconvfilter_rn3ry0.pt', 'random_01_run1_allnoise_correctbyfcndaefilter_rn3ry0.pt', 'random_01_run1_allnoise_correctbyconvlstmfilter_rn3ry0.pt', 'random_01_run1_allnoise_correctbyconvlstm2filter_rn3ry0.pt', 'random_01_run1_allnoise_correctbyconvlstm3filter_rn3ry0.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter_rn3ry0.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn3ry0.pt', 'random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn3ry0.pt', 'random_01_run1_all_rn4ry0.pt', 'random_01_run1_noneut_rn4ry0.pt', 'random_01_run1_high_rn4ry0.pt', 'random_01_run1_all_corrected_rn4ry0.pt', 'random_01_run1_all_corrected_noise_rn4ry0.pt', 'random_01_run1_scope_corrected_noise_rn4ry0.pt', 'random_01_run1_noneut_corrected_rn4ry0.pt', 'random_01_run1_allnoise_correctedhwgt_rn4ry0.pt', 'random_01_run1_allnoise_correctedscore_rn4ry0.pt', 'random_01_run1_allnoise_correctbydeepconvfilter_rn4ry0.pt', 'random_01_run1_allnoise_correctbyfcndaefilter_rn4ry0.pt', 'random_01_run1_allnoise_correctbyconvlstmfilter_rn4ry0.pt', 'random_01_run1_allnoise_correctbyconvlstm2filter_rn4ry0.pt', 'random_01_run1_allnoise_correctbyconvlstm3filter_rn4ry0.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter_rn4ry0.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn4ry0.pt', 'random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn4ry0.pt', 'random_01_run1_all_rn5ry0.pt', 'random_01_run1_noneut_rn5ry0.pt', 'random_01_run1_high_rn5ry0.pt', 'random_01_run1_all_corrected_rn5ry0.pt', 'random_01_run1_all_corrected_noise_rn5ry0.pt', 'random_01_run1_scope_corrected_noise_rn5ry0.pt', 'random_01_run1_noneut_corrected_rn5ry0.pt', 'random_01_run1_allnoise_correctedhwgt_rn5ry0.pt', 'random_01_run1_allnoise_correctedscore_rn5ry0.pt', 'random_01_run1_allnoise_correctbydeepconvfilter_rn5ry0.pt', 'random_01_run1_allnoise_correctbyfcndaefilter_rn5ry0.pt', 'random_01_run1_allnoise_correctbyconvlstmfilter_rn5ry0.pt', 'random_01_run1_allnoise_correctbyconvlstm2filter_rn5ry0.pt', 'random_01_run1_allnoise_correctbyconvlstm3filter_rn5ry0.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter_rn5ry0.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn5ry0.pt', 'random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn5ry0.pt', 'random_01_run1_all_rn6ry0.pt', 'random_01_run1_noneut_rn6ry0.pt', 'random_01_run1_high_rn6ry0.pt', 'random_01_run1_all_corrected_rn6ry0.pt', 'random_01_run1_all_corrected_noise_rn6ry0.pt', 'random_01_run1_scope_corrected_noise_rn6ry0.pt', 'random_01_run1_noneut_corrected_rn6ry0.pt', 'random_01_run1_allnoise_correctedhwgt_rn6ry0.pt', 'random_01_run1_allnoise_correctedscore_rn6ry0.pt', 'random_01_run1_allnoise_correctbydeepconvfilter_rn6ry0.pt', 'random_01_run1_allnoise_correctbyfcndaefilter_rn6ry0.pt', 'random_01_run1_allnoise_correctbyconvlstmfilter_rn6ry0.pt', 'random_01_run1_allnoise_correctbyconvlstm2filter_rn6ry0.pt', 'random_01_run1_allnoise_correctbyconvlstm3filter_rn6ry0.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter_rn6ry0.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn6ry0.pt', 'random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn6ry0.pt', 'random_01_run1_all_rn7ry0.pt', 'random_01_run1_noneut_rn7ry0.pt', 'random_01_run1_high_rn7ry0.pt', 'random_01_run1_all_corrected_rn7ry0.pt', 'random_01_run1_all_corrected_noise_rn7ry0.pt', 'random_01_run1_scope_corrected_noise_rn7ry0.pt', 'random_01_run1_noneut_corrected_rn7ry0.pt', 'random_01_run1_allnoise_correctedhwgt_rn7ry0.pt', 'random_01_run1_allnoise_correctedscore_rn7ry0.pt', 'random_01_run1_allnoise_correctbydeepconvfilter_rn7ry0.pt', 'random_01_run1_allnoise_correctbyfcndaefilter_rn7ry0.pt', 'random_01_run1_allnoise_correctbyconvlstmfilter_rn7ry0.pt', 'random_01_run1_allnoise_correctbyconvlstm2filter_rn7ry0.pt', 'random_01_run1_allnoise_correctbyconvlstm3filter_rn7ry0.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter_rn7ry0.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn7ry0.pt', 'random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn7ry0.pt', 'random_01_run1_all_rn8ry0.pt', 'random_01_run1_noneut_rn8ry0.pt', 'random_01_run1_high_rn8ry0.pt', 'random_01_run1_all_corrected_rn8ry0.pt', 'random_01_run1_all_corrected_noise_rn8ry0.pt', 'random_01_run1_scope_corrected_noise_rn8ry0.pt', 'random_01_run1_noneut_corrected_rn8ry0.pt', 'random_01_run1_allnoise_correctedhwgt_rn8ry0.pt', 'random_01_run1_allnoise_correctedscore_rn8ry0.pt', 'random_01_run1_allnoise_correctbydeepconvfilter_rn8ry0.pt', 'random_01_run1_allnoise_correctbyfcndaefilter_rn8ry0.pt', 'random_01_run1_allnoise_correctbyconvlstmfilter_rn8ry0.pt', 'random_01_run1_allnoise_correctbyconvlstm2filter_rn8ry0.pt', 'random_01_run1_allnoise_correctbyconvlstm3filter_rn8ry0.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter_rn8ry0.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn8ry0.pt', 'random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn8ry0.pt', 'random_01_run1_all_rn9ry0.pt', 'random_01_run1_noneut_rn9ry0.pt', 'random_01_run1_high_rn9ry0.pt', 'random_01_run1_all_corrected_rn9ry0.pt', 'random_01_run1_all_corrected_noise_rn9ry0.pt', 'random_01_run1_scope_corrected_noise_rn9ry0.pt', 'random_01_run1_noneut_corrected_rn9ry0.pt', 'random_01_run1_allnoise_correctedhwgt_rn9ry0.pt', 'random_01_run1_allnoise_correctedscore_rn9ry0.pt', 'random_01_run1_allnoise_correctbydeepconvfilter_rn9ry0.pt', 'random_01_run1_allnoise_correctbyfcndaefilter_rn9ry0.pt', 'random_01_run1_allnoise_correctbyconvlstmfilter_rn9ry0.pt', 'random_01_run1_allnoise_correctbyconvlstm2filter_rn9ry0.pt', 'random_01_run1_allnoise_correctbyconvlstm3filter_rn9ry0.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter_rn9ry0.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn9ry0.pt', 'random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn9ry0.pt', 'random_01_run1_all_rn0ry0.pt', 'random_01_run1_noneut_rn0ry0.pt', 'random_01_run1_high_rn0ry0.pt', 'random_01_run1_all_corrected_rn0ry0.pt', 'random_01_run1_all_corrected_noise_rn0ry0.pt', 'random_01_run1_scope_corrected_noise_rn0ry0.pt', 'random_01_run1_noneut_corrected_rn0ry0.pt', 'random_01_run1_allnoise_correctedhwgt_rn0ry0.pt', 'random_01_run1_allnoise_correctedscore_rn0ry0.pt', 'random_01_run1_allnoise_correctbydeepconvfilter_rn0ry0.pt', 'random_01_run1_allnoise_correctbyfcndaefilter_rn0ry0.pt', 'random_01_run1_allnoise_correctbyconvlstmfilter_rn0ry0.pt', 'random_01_run1_allnoise_correctbyconvlstm2filter_rn0ry0.pt', 'random_01_run1_allnoise_correctbyconvlstm3filter_rn0ry0.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry0.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry0.pt', 'random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry0.pt', 'random_01_run1_all_rn0ry2.pt', 'random_01_run1_noneut_rn0ry2.pt', 'random_01_run1_high_rn0ry2.pt', 'random_01_run1_all_corrected_rn0ry2.pt', 'random_01_run1_all_corrected_noise_rn0ry2.pt', 'random_01_run1_scope_corrected_noise_rn0ry2.pt', 'random_01_run1_noneut_corrected_rn0ry2.pt', 'random_01_run1_allnoise_correctedhwgt_rn0ry2.pt', 'random_01_run1_allnoise_correctedscore_rn0ry2.pt', 'random_01_run1_allnoise_correctbydeepconvfilter_rn0ry2.pt', 'random_01_run1_allnoise_correctbyfcndaefilter_rn0ry2.pt', 'random_01_run1_allnoise_correctbyconvlstmfilter_rn0ry2.pt', 'random_01_run1_allnoise_correctbyconvlstm2filter_rn0ry2.pt', 'random_01_run1_allnoise_correctbyconvlstm3filter_rn0ry2.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry2.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry2.pt', 'random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry2.pt', 'random_01_run1_all_rn0ry3.pt', 'random_01_run1_noneut_rn0ry3.pt', 'random_01_run1_high_rn0ry3.pt', 'random_01_run1_all_corrected_rn0ry3.pt', 'random_01_run1_all_corrected_noise_rn0ry3.pt', 'random_01_run1_scope_corrected_noise_rn0ry3.pt', 'random_01_run1_noneut_corrected_rn0ry3.pt', 'random_01_run1_allnoise_correctedhwgt_rn0ry3.pt', 'random_01_run1_allnoise_correctedscore_rn0ry3.pt', 'random_01_run1_allnoise_correctbydeepconvfilter_rn0ry3.pt', 'random_01_run1_allnoise_correctbyfcndaefilter_rn0ry3.pt', 'random_01_run1_allnoise_correctbyconvlstmfilter_rn0ry3.pt', 'random_01_run1_allnoise_correctbyconvlstm2filter_rn0ry3.pt', 'random_01_run1_allnoise_correctbyconvlstm3filter_rn0ry3.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry3.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry3.pt', 'random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry3.pt', 'random_01_run1_all_rn0ry4.pt', 'random_01_run1_noneut_rn0ry4.pt', 'random_01_run1_high_rn0ry4.pt', 'random_01_run1_all_corrected_rn0ry4.pt', 'random_01_run1_all_corrected_noise_rn0ry4.pt', 'random_01_run1_scope_corrected_noise_rn0ry4.pt', 'random_01_run1_noneut_corrected_rn0ry4.pt', 'random_01_run1_allnoise_correctedhwgt_rn0ry4.pt', 'random_01_run1_allnoise_correctedscore_rn0ry4.pt', 'random_01_run1_allnoise_correctbydeepconvfilter_rn0ry4.pt', 'random_01_run1_allnoise_correctbyfcndaefilter_rn0ry4.pt', 'random_01_run1_allnoise_correctbyconvlstmfilter_rn0ry4.pt', 'random_01_run1_allnoise_correctbyconvlstm2filter_rn0ry4.pt', 'random_01_run1_allnoise_correctbyconvlstm3filter_rn0ry4.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry4.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry4.pt', 'random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry4.pt', 'random_01_run1_all_rn0ry5.pt', 'random_01_run1_noneut_rn0ry5.pt', 'random_01_run1_high_rn0ry5.pt', 'random_01_run1_all_corrected_rn0ry5.pt', 'random_01_run1_all_corrected_noise_rn0ry5.pt', 'random_01_run1_scope_corrected_noise_rn0ry5.pt', 'random_01_run1_noneut_corrected_rn0ry5.pt', 'random_01_run1_allnoise_correctedhwgt_rn0ry5.pt', 'random_01_run1_allnoise_correctedscore_rn0ry5.pt', 'random_01_run1_allnoise_correctbydeepconvfilter_rn0ry5.pt', 'random_01_run1_allnoise_correctbyfcndaefilter_rn0ry5.pt', 'random_01_run1_allnoise_correctbyconvlstmfilter_rn0ry5.pt', 'random_01_run1_allnoise_correctbyconvlstm2filter_rn0ry5.pt', 'random_01_run1_allnoise_correctbyconvlstm3filter_rn0ry5.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry5.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry5.pt', 'random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry5.pt', 'random_01_run1_all_rn0ry6.pt', 'random_01_run1_noneut_rn0ry6.pt', 'random_01_run1_high_rn0ry6.pt', 'random_01_run1_all_corrected_rn0ry6.pt', 'random_01_run1_all_corrected_noise_rn0ry6.pt', 'random_01_run1_scope_corrected_noise_rn0ry6.pt', 'random_01_run1_noneut_corrected_rn0ry6.pt', 'random_01_run1_allnoise_correctedhwgt_rn0ry6.pt', 'random_01_run1_allnoise_correctedscore_rn0ry6.pt', 'random_01_run1_allnoise_correctbydeepconvfilter_rn0ry6.pt', 'random_01_run1_allnoise_correctbyfcndaefilter_rn0ry6.pt', 'random_01_run1_allnoise_correctbyconvlstmfilter_rn0ry6.pt', 'random_01_run1_allnoise_correctbyconvlstm2filter_rn0ry6.pt', 'random_01_run1_allnoise_correctbyconvlstm3filter_rn0ry6.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry6.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry6.pt', 'random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry6.pt', 'random_01_run1_all_rn0ry7.pt', 'random_01_run1_noneut_rn0ry7.pt', 'random_01_run1_high_rn0ry7.pt', 'random_01_run1_all_corrected_rn0ry7.pt', 'random_01_run1_all_corrected_noise_rn0ry7.pt', 'random_01_run1_scope_corrected_noise_rn0ry7.pt', 'random_01_run1_noneut_corrected_rn0ry7.pt', 'random_01_run1_allnoise_correctedhwgt_rn0ry7.pt', 'random_01_run1_allnoise_correctedscore_rn0ry7.pt', 'random_01_run1_allnoise_correctbydeepconvfilter_rn0ry7.pt', 'random_01_run1_allnoise_correctbyfcndaefilter_rn0ry7.pt', 'random_01_run1_allnoise_correctbyconvlstmfilter_rn0ry7.pt', 'random_01_run1_allnoise_correctbyconvlstm2filter_rn0ry7.pt', 'random_01_run1_allnoise_correctbyconvlstm3filter_rn0ry7.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry7.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry7.pt', 'random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry7.pt', 'random_01_run1_all_rn0ry8.pt', 'random_01_run1_noneut_rn0ry8.pt', 'random_01_run1_high_rn0ry8.pt', 'random_01_run1_all_corrected_rn0ry8.pt', 'random_01_run1_all_corrected_noise_rn0ry8.pt', 'random_01_run1_scope_corrected_noise_rn0ry8.pt', 'random_01_run1_noneut_corrected_rn0ry8.pt', 'random_01_run1_allnoise_correctedhwgt_rn0ry8.pt', 'random_01_run1_allnoise_correctedscore_rn0ry8.pt', 'random_01_run1_allnoise_correctbydeepconvfilter_rn0ry8.pt', 'random_01_run1_allnoise_correctbyfcndaefilter_rn0ry8.pt', 'random_01_run1_allnoise_correctbyconvlstmfilter_rn0ry8.pt', 'random_01_run1_allnoise_correctbyconvlstm2filter_rn0ry8.pt', 'random_01_run1_allnoise_correctbyconvlstm3filter_rn0ry8.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry8.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry8.pt', 'random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry8.pt', 'random_01_run1_all_rn0ry9.pt', 'random_01_run1_noneut_rn0ry9.pt', 'random_01_run1_high_rn0ry9.pt', 'random_01_run1_all_corrected_rn0ry9.pt', 'random_01_run1_all_corrected_noise_rn0ry9.pt', 'random_01_run1_scope_corrected_noise_rn0ry9.pt', 'random_01_run1_noneut_corrected_rn0ry9.pt', 'random_01_run1_allnoise_correctedhwgt_rn0ry9.pt', 'random_01_run1_allnoise_correctedscore_rn0ry9.pt', 'random_01_run1_allnoise_correctbydeepconvfilter_rn0ry9.pt', 'random_01_run1_allnoise_correctbyfcndaefilter_rn0ry9.pt', 'random_01_run1_allnoise_correctbyconvlstmfilter_rn0ry9.pt', 'random_01_run1_allnoise_correctbyconvlstm2filter_rn0ry9.pt', 'random_01_run1_allnoise_correctbyconvlstm3filter_rn0ry9.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry9.pt', 'random_01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry9.pt', 'random_01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry9.pt', 'DeepConvLSTMA_statedict.pt', 'recorded_inc_01.pt', 'rec01_run1_all.pt', 'rec01_run1_scope.pt', 'rec01_run1_noneut.pt', 'rec01_run1_high.pt', 'rec01_run1_inchc.pt', 'rec01_run1_all_corrected.pt', 'rec01_run1_all_corrected_noise.pt', 'rec01_run1_scope_corrected_noise.pt', 'rec01_run1_scope_corrected.pt', 'rec01_run1_noneut_corrected.pt', 'rec01_run1_correctedhwgt.pt', 'rec01_run1_noneut_correctedhwgt.pt', 'rec01_run1_inchc_correctedhwgt.pt', 'rec01_run1_synthetic.pt', 'rec01_run1_allnoise_correctedhwgt.pt', 'rec01_run1_allnoise_correctedhwgt_scope.pt', 'rec01_run1_allnoise_corrected.pt', 'rec01_run1_allnoise_correctedscore.pt', 'rec01_run1_allnoise_correctedscore_flatten.pt', 'rec01_run1_allnoise_correctbydeepconvfilter.pt', 'rec01_run1_allnoise_correctbyfcndaefilter.pt', 'rec01_run1_allnoise_correctbyfcndaefiltertest.pt', 'rec01_run1_allnoise_correctbyconvlstmfilter.pt', 'rec01_run1_allnoise_correctbyconvlstm2filter.pt', 'rec01_run1_allnoise_correctbyconvlstm3filter.pt', 'rec01_run1_alldeepconv_correctbyconvlstm3filter.pt', 'rec01_run1_alldeepconv_correctbyconvlstm3filter2.pt', 'rec01_run1_alldeepconv_correctbyconvlstm3filter3.pt', 'rec01_run1_alldeepconv_correctbyconvlstm3filter4.pt', 'rec01_run1_alldeepconv_correctbyconvlstm3filter5.pt', 'rec01_run1_alldeepconv_correctbyconvlstm3filter6.pt', 'rec01_run1_alldeepconv_correctbyconvlstm2filter6.pt', 'recorded_inc_02.pt', 'rec02_run1_all.pt', 'rec02_run1_scope.pt', 'rec02_run1_noneut.pt', 'rec02_run1_high.pt', 'rec02_run1_inchc.pt', 'rec02_run1_all_corrected.pt', 'rec02_run1_all_corrected_noise.pt', 'rec02_run1_scope_corrected_noise.pt', 'rec02_run1_scope_corrected.pt', 'rec02_run1_noneut_corrected.pt', 'rec02_run1_correctedhwgt.pt', 'rec02_run1_noneut_correctedhwgt.pt', 'rec02_run1_inchc_correctedhwgt.pt', 'rec02_run1_synthetic.pt', 'rec02_run1_allnoise_correctedhwgt.pt', 'rec02_run1_allnoise_correctedhwgt_scope.pt', 'rec02_run1_allnoise_corrected.pt', 'rec02_run1_allnoise_correctedscore.pt', 'rec02_run1_allnoise_correctedscore_flatten.pt', 'rec02_run1_allnoise_correctbydeepconvfilter.pt', 'rec02_run1_allnoise_correctbyfcndaefilter.pt', 'rec02_run1_allnoise_correctbyfcndaefiltertest.pt', 'rec02_run1_allnoise_correctbyconvlstmfilter.pt', 'rec02_run1_allnoise_correctbyconvlstm2filter.pt', 'rec02_run1_allnoise_correctbyconvlstm3filter.pt', 'rec02_run1_alldeepconv_correctbyconvlstm3filter.pt', 'rec02_run1_alldeepconv_correctbyconvlstm3filter2.pt', 'rec02_run1_alldeepconv_correctbyconvlstm3filter3.pt', 'rec02_run1_alldeepconv_correctbyconvlstm3filter4.pt', 'rec02_run1_alldeepconv_correctbyconvlstm3filter5.pt', 'rec02_run1_alldeepconv_correctbyconvlstm3filter6.pt', 'rec02_run1_alldeepconv_correctbyconvlstm2filter6.pt', 'random_rec01_run1_all_rn2ry0.pt', 'random_rec01_run1_noneut_rn2ry0.pt', 'random_rec01_run1_high_rn2ry0.pt', 'random_rec01_run1_all_corrected_rn2ry0.pt', 'random_rec01_run1_all_corrected_noise_rn2ry0.pt', 'random_rec01_run1_scope_corrected_noise_rn2ry0.pt', 'random_rec01_run1_noneut_corrected_rn2ry0.pt', 'random_rec01_run1_allnoise_correctedhwgt_rn2ry0.pt', 'random_rec01_run1_allnoise_correctedscore_rn2ry0.pt', 'random_rec01_run1_allnoise_correctbydeepconvfilter_rn2ry0.pt', 'random_rec01_run1_allnoise_correctbyfcndaefilter_rn2ry0.pt', 'random_rec01_run1_allnoise_correctbyconvlstmfilter_rn2ry0.pt', 'random_rec01_run1_allnoise_correctbyconvlstm2filter_rn2ry0.pt', 'random_rec01_run1_allnoise_correctbyconvlstm3filter_rn2ry0.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter_rn2ry0.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter6_rn2ry0.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm2filter6_rn2ry0.pt', 'random_rec01_run1_all_rn3ry0.pt', 'random_rec01_run1_noneut_rn3ry0.pt', 'random_rec01_run1_high_rn3ry0.pt', 'random_rec01_run1_all_corrected_rn3ry0.pt', 'random_rec01_run1_all_corrected_noise_rn3ry0.pt', 'random_rec01_run1_scope_corrected_noise_rn3ry0.pt', 'random_rec01_run1_noneut_corrected_rn3ry0.pt', 'random_rec01_run1_allnoise_correctedhwgt_rn3ry0.pt', 'random_rec01_run1_allnoise_correctedscore_rn3ry0.pt', 'random_rec01_run1_allnoise_correctbydeepconvfilter_rn3ry0.pt', 'random_rec01_run1_allnoise_correctbyfcndaefilter_rn3ry0.pt', 'random_rec01_run1_allnoise_correctbyconvlstmfilter_rn3ry0.pt', 'random_rec01_run1_allnoise_correctbyconvlstm2filter_rn3ry0.pt', 'random_rec01_run1_allnoise_correctbyconvlstm3filter_rn3ry0.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter_rn3ry0.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter6_rn3ry0.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm2filter6_rn3ry0.pt', 'random_rec01_run1_all_rn4ry0.pt', 'random_rec01_run1_noneut_rn4ry0.pt', 'random_rec01_run1_high_rn4ry0.pt', 'random_rec01_run1_all_corrected_rn4ry0.pt', 'random_rec01_run1_all_corrected_noise_rn4ry0.pt', 'random_rec01_run1_scope_corrected_noise_rn4ry0.pt', 'random_rec01_run1_noneut_corrected_rn4ry0.pt', 'random_rec01_run1_allnoise_correctedhwgt_rn4ry0.pt', 'random_rec01_run1_allnoise_correctedscore_rn4ry0.pt', 'random_rec01_run1_allnoise_correctbydeepconvfilter_rn4ry0.pt', 'random_rec01_run1_allnoise_correctbyfcndaefilter_rn4ry0.pt', 'random_rec01_run1_allnoise_correctbyconvlstmfilter_rn4ry0.pt', 'random_rec01_run1_allnoise_correctbyconvlstm2filter_rn4ry0.pt', 'random_rec01_run1_allnoise_correctbyconvlstm3filter_rn4ry0.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter_rn4ry0.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter6_rn4ry0.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm2filter6_rn4ry0.pt', 'random_rec01_run1_all_rn5ry0.pt', 'random_rec01_run1_noneut_rn5ry0.pt', 'random_rec01_run1_high_rn5ry0.pt', 'random_rec01_run1_all_corrected_rn5ry0.pt', 'random_rec01_run1_all_corrected_noise_rn5ry0.pt', 'random_rec01_run1_scope_corrected_noise_rn5ry0.pt', 'random_rec01_run1_noneut_corrected_rn5ry0.pt', 'random_rec01_run1_allnoise_correctedhwgt_rn5ry0.pt', 'random_rec01_run1_allnoise_correctedscore_rn5ry0.pt', 'random_rec01_run1_allnoise_correctbydeepconvfilter_rn5ry0.pt', 'random_rec01_run1_allnoise_correctbyfcndaefilter_rn5ry0.pt', 'random_rec01_run1_allnoise_correctbyconvlstmfilter_rn5ry0.pt', 'random_rec01_run1_allnoise_correctbyconvlstm2filter_rn5ry0.pt', 'random_rec01_run1_allnoise_correctbyconvlstm3filter_rn5ry0.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter_rn5ry0.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter6_rn5ry0.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm2filter6_rn5ry0.pt', 'random_rec01_run1_all_rn6ry0.pt', 'random_rec01_run1_noneut_rn6ry0.pt', 'random_rec01_run1_high_rn6ry0.pt', 'random_rec01_run1_all_corrected_rn6ry0.pt', 'random_rec01_run1_all_corrected_noise_rn6ry0.pt', 'random_rec01_run1_scope_corrected_noise_rn6ry0.pt', 'random_rec01_run1_noneut_corrected_rn6ry0.pt', 'random_rec01_run1_allnoise_correctedhwgt_rn6ry0.pt', 'random_rec01_run1_allnoise_correctedscore_rn6ry0.pt', 'random_rec01_run1_allnoise_correctbydeepconvfilter_rn6ry0.pt', 'random_rec01_run1_allnoise_correctbyfcndaefilter_rn6ry0.pt', 'random_rec01_run1_allnoise_correctbyconvlstmfilter_rn6ry0.pt', 'random_rec01_run1_allnoise_correctbyconvlstm2filter_rn6ry0.pt', 'random_rec01_run1_allnoise_correctbyconvlstm3filter_rn6ry0.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter_rn6ry0.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter6_rn6ry0.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm2filter6_rn6ry0.pt', 'random_rec01_run1_all_rn7ry0.pt', 'random_rec01_run1_noneut_rn7ry0.pt', 'random_rec01_run1_high_rn7ry0.pt', 'random_rec01_run1_all_corrected_rn7ry0.pt', 'random_rec01_run1_all_corrected_noise_rn7ry0.pt', 'random_rec01_run1_scope_corrected_noise_rn7ry0.pt', 'random_rec01_run1_noneut_corrected_rn7ry0.pt', 'random_rec01_run1_allnoise_correctedhwgt_rn7ry0.pt', 'random_rec01_run1_allnoise_correctedscore_rn7ry0.pt', 'random_rec01_run1_allnoise_correctbydeepconvfilter_rn7ry0.pt', 'random_rec01_run1_allnoise_correctbyfcndaefilter_rn7ry0.pt', 'random_rec01_run1_allnoise_correctbyconvlstmfilter_rn7ry0.pt', 'random_rec01_run1_allnoise_correctbyconvlstm2filter_rn7ry0.pt', 'random_rec01_run1_allnoise_correctbyconvlstm3filter_rn7ry0.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter_rn7ry0.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter6_rn7ry0.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm2filter6_rn7ry0.pt', 'random_rec01_run1_all_rn8ry0.pt', 'random_rec01_run1_noneut_rn8ry0.pt', 'random_rec01_run1_high_rn8ry0.pt', 'random_rec01_run1_all_corrected_rn8ry0.pt', 'random_rec01_run1_all_corrected_noise_rn8ry0.pt', 'random_rec01_run1_scope_corrected_noise_rn8ry0.pt', 'random_rec01_run1_noneut_corrected_rn8ry0.pt', 'random_rec01_run1_allnoise_correctedhwgt_rn8ry0.pt', 'random_rec01_run1_allnoise_correctedscore_rn8ry0.pt', 'random_rec01_run1_allnoise_correctbydeepconvfilter_rn8ry0.pt', 'random_rec01_run1_allnoise_correctbyfcndaefilter_rn8ry0.pt', 'random_rec01_run1_allnoise_correctbyconvlstmfilter_rn8ry0.pt', 'random_rec01_run1_allnoise_correctbyconvlstm2filter_rn8ry0.pt', 'random_rec01_run1_allnoise_correctbyconvlstm3filter_rn8ry0.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter_rn8ry0.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter6_rn8ry0.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm2filter6_rn8ry0.pt', 'random_rec01_run1_all_rn9ry0.pt', 'random_rec01_run1_noneut_rn9ry0.pt', 'random_rec01_run1_high_rn9ry0.pt', 'random_rec01_run1_all_corrected_rn9ry0.pt', 'random_rec01_run1_all_corrected_noise_rn9ry0.pt', 'random_rec01_run1_scope_corrected_noise_rn9ry0.pt', 'random_rec01_run1_noneut_corrected_rn9ry0.pt', 'random_rec01_run1_allnoise_correctedhwgt_rn9ry0.pt', 'random_rec01_run1_allnoise_correctedscore_rn9ry0.pt', 'random_rec01_run1_allnoise_correctbydeepconvfilter_rn9ry0.pt', 'random_rec01_run1_allnoise_correctbyfcndaefilter_rn9ry0.pt', 'random_rec01_run1_allnoise_correctbyconvlstmfilter_rn9ry0.pt', 'random_rec01_run1_allnoise_correctbyconvlstm2filter_rn9ry0.pt', 'random_rec01_run1_allnoise_correctbyconvlstm3filter_rn9ry0.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter_rn9ry0.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter6_rn9ry0.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm2filter6_rn9ry0.pt', 'random_rec01_run1_all_rn0ry0.pt', 'random_rec01_run1_noneut_rn0ry0.pt', 'random_rec01_run1_high_rn0ry0.pt', 'random_rec01_run1_all_corrected_rn0ry0.pt', 'random_rec01_run1_all_corrected_noise_rn0ry0.pt', 'random_rec01_run1_scope_corrected_noise_rn0ry0.pt', 'random_rec01_run1_noneut_corrected_rn0ry0.pt', 'random_rec01_run1_allnoise_correctedhwgt_rn0ry0.pt', 'random_rec01_run1_allnoise_correctedscore_rn0ry0.pt', 'random_rec01_run1_allnoise_correctbydeepconvfilter_rn0ry0.pt', 'random_rec01_run1_allnoise_correctbyfcndaefilter_rn0ry0.pt', 'random_rec01_run1_allnoise_correctbyconvlstmfilter_rn0ry0.pt', 'random_rec01_run1_allnoise_correctbyconvlstm2filter_rn0ry0.pt', 'random_rec01_run1_allnoise_correctbyconvlstm3filter_rn0ry0.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry0.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry0.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry0.pt', 'random_rec01_run1_all_rn0ry2.pt', 'random_rec01_run1_noneut_rn0ry2.pt', 'random_rec01_run1_high_rn0ry2.pt', 'random_rec01_run1_all_corrected_rn0ry2.pt', 'random_rec01_run1_all_corrected_noise_rn0ry2.pt', 'random_rec01_run1_scope_corrected_noise_rn0ry2.pt', 'random_rec01_run1_noneut_corrected_rn0ry2.pt', 'random_rec01_run1_allnoise_correctedhwgt_rn0ry2.pt', 'random_rec01_run1_allnoise_correctedscore_rn0ry2.pt', 'random_rec01_run1_allnoise_correctbydeepconvfilter_rn0ry2.pt', 'random_rec01_run1_allnoise_correctbyfcndaefilter_rn0ry2.pt', 'random_rec01_run1_allnoise_correctbyconvlstmfilter_rn0ry2.pt', 'random_rec01_run1_allnoise_correctbyconvlstm2filter_rn0ry2.pt', 'random_rec01_run1_allnoise_correctbyconvlstm3filter_rn0ry2.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry2.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry2.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry2.pt', 'random_rec01_run1_all_rn0ry3.pt', 'random_rec01_run1_noneut_rn0ry3.pt', 'random_rec01_run1_high_rn0ry3.pt', 'random_rec01_run1_all_corrected_rn0ry3.pt', 'random_rec01_run1_all_corrected_noise_rn0ry3.pt', 'random_rec01_run1_scope_corrected_noise_rn0ry3.pt', 'random_rec01_run1_noneut_corrected_rn0ry3.pt', 'random_rec01_run1_allnoise_correctedhwgt_rn0ry3.pt', 'random_rec01_run1_allnoise_correctedscore_rn0ry3.pt', 'random_rec01_run1_allnoise_correctbydeepconvfilter_rn0ry3.pt', 'random_rec01_run1_allnoise_correctbyfcndaefilter_rn0ry3.pt', 'random_rec01_run1_allnoise_correctbyconvlstmfilter_rn0ry3.pt', 'random_rec01_run1_allnoise_correctbyconvlstm2filter_rn0ry3.pt', 'random_rec01_run1_allnoise_correctbyconvlstm3filter_rn0ry3.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry3.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry3.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry3.pt', 'random_rec01_run1_all_rn0ry4.pt', 'random_rec01_run1_noneut_rn0ry4.pt', 'random_rec01_run1_high_rn0ry4.pt', 'random_rec01_run1_all_corrected_rn0ry4.pt', 'random_rec01_run1_all_corrected_noise_rn0ry4.pt', 'random_rec01_run1_scope_corrected_noise_rn0ry4.pt', 'random_rec01_run1_noneut_corrected_rn0ry4.pt', 'random_rec01_run1_allnoise_correctedhwgt_rn0ry4.pt', 'random_rec01_run1_allnoise_correctedscore_rn0ry4.pt', 'random_rec01_run1_allnoise_correctbydeepconvfilter_rn0ry4.pt', 'random_rec01_run1_allnoise_correctbyfcndaefilter_rn0ry4.pt', 'random_rec01_run1_allnoise_correctbyconvlstmfilter_rn0ry4.pt', 'random_rec01_run1_allnoise_correctbyconvlstm2filter_rn0ry4.pt', 'random_rec01_run1_allnoise_correctbyconvlstm3filter_rn0ry4.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry4.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry4.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry4.pt', 'random_rec01_run1_all_rn0ry5.pt', 'random_rec01_run1_noneut_rn0ry5.pt', 'random_rec01_run1_high_rn0ry5.pt', 'random_rec01_run1_all_corrected_rn0ry5.pt', 'random_rec01_run1_all_corrected_noise_rn0ry5.pt', 'random_rec01_run1_scope_corrected_noise_rn0ry5.pt', 'random_rec01_run1_noneut_corrected_rn0ry5.pt', 'random_rec01_run1_allnoise_correctedhwgt_rn0ry5.pt', 'random_rec01_run1_allnoise_correctedscore_rn0ry5.pt', 'random_rec01_run1_allnoise_correctbydeepconvfilter_rn0ry5.pt', 'random_rec01_run1_allnoise_correctbyfcndaefilter_rn0ry5.pt', 'random_rec01_run1_allnoise_correctbyconvlstmfilter_rn0ry5.pt', 'random_rec01_run1_allnoise_correctbyconvlstm2filter_rn0ry5.pt', 'random_rec01_run1_allnoise_correctbyconvlstm3filter_rn0ry5.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry5.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry5.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry5.pt', 'random_rec01_run1_all_rn0ry6.pt', 'random_rec01_run1_noneut_rn0ry6.pt', 'random_rec01_run1_high_rn0ry6.pt', 'random_rec01_run1_all_corrected_rn0ry6.pt', 'random_rec01_run1_all_corrected_noise_rn0ry6.pt', 'random_rec01_run1_scope_corrected_noise_rn0ry6.pt', 'random_rec01_run1_noneut_corrected_rn0ry6.pt', 'random_rec01_run1_allnoise_correctedhwgt_rn0ry6.pt', 'random_rec01_run1_allnoise_correctedscore_rn0ry6.pt', 'random_rec01_run1_allnoise_correctbydeepconvfilter_rn0ry6.pt', 'random_rec01_run1_allnoise_correctbyfcndaefilter_rn0ry6.pt', 'random_rec01_run1_allnoise_correctbyconvlstmfilter_rn0ry6.pt', 'random_rec01_run1_allnoise_correctbyconvlstm2filter_rn0ry6.pt', 'random_rec01_run1_allnoise_correctbyconvlstm3filter_rn0ry6.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry6.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry6.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry6.pt', 'random_rec01_run1_all_rn0ry7.pt', 'random_rec01_run1_noneut_rn0ry7.pt', 'random_rec01_run1_high_rn0ry7.pt', 'random_rec01_run1_all_corrected_rn0ry7.pt', 'random_rec01_run1_all_corrected_noise_rn0ry7.pt', 'random_rec01_run1_scope_corrected_noise_rn0ry7.pt', 'random_rec01_run1_noneut_corrected_rn0ry7.pt', 'random_rec01_run1_allnoise_correctedhwgt_rn0ry7.pt', 'random_rec01_run1_allnoise_correctedscore_rn0ry7.pt', 'random_rec01_run1_allnoise_correctbydeepconvfilter_rn0ry7.pt', 'random_rec01_run1_allnoise_correctbyfcndaefilter_rn0ry7.pt', 'random_rec01_run1_allnoise_correctbyconvlstmfilter_rn0ry7.pt', 'random_rec01_run1_allnoise_correctbyconvlstm2filter_rn0ry7.pt', 'random_rec01_run1_allnoise_correctbyconvlstm3filter_rn0ry7.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry7.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry7.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry7.pt', 'random_rec01_run1_all_rn0ry8.pt', 'random_rec01_run1_noneut_rn0ry8.pt', 'random_rec01_run1_high_rn0ry8.pt', 'random_rec01_run1_all_corrected_rn0ry8.pt', 'random_rec01_run1_all_corrected_noise_rn0ry8.pt', 'random_rec01_run1_scope_corrected_noise_rn0ry8.pt', 'random_rec01_run1_noneut_corrected_rn0ry8.pt', 'random_rec01_run1_allnoise_correctedhwgt_rn0ry8.pt', 'random_rec01_run1_allnoise_correctedscore_rn0ry8.pt', 'random_rec01_run1_allnoise_correctbydeepconvfilter_rn0ry8.pt', 'random_rec01_run1_allnoise_correctbyfcndaefilter_rn0ry8.pt', 'random_rec01_run1_allnoise_correctbyconvlstmfilter_rn0ry8.pt', 'random_rec01_run1_allnoise_correctbyconvlstm2filter_rn0ry8.pt', 'random_rec01_run1_allnoise_correctbyconvlstm3filter_rn0ry8.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry8.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry8.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry8.pt', 'random_rec01_run1_all_rn0ry9.pt', 'random_rec01_run1_noneut_rn0ry9.pt', 'random_rec01_run1_high_rn0ry9.pt', 'random_rec01_run1_all_corrected_rn0ry9.pt', 'random_rec01_run1_all_corrected_noise_rn0ry9.pt', 'random_rec01_run1_scope_corrected_noise_rn0ry9.pt', 'random_rec01_run1_noneut_corrected_rn0ry9.pt', 'random_rec01_run1_allnoise_correctedhwgt_rn0ry9.pt', 'random_rec01_run1_allnoise_correctedscore_rn0ry9.pt', 'random_rec01_run1_allnoise_correctbydeepconvfilter_rn0ry9.pt', 'random_rec01_run1_allnoise_correctbyfcndaefilter_rn0ry9.pt', 'random_rec01_run1_allnoise_correctbyconvlstmfilter_rn0ry9.pt', 'random_rec01_run1_allnoise_correctbyconvlstm2filter_rn0ry9.pt', 'random_rec01_run1_allnoise_correctbyconvlstm3filter_rn0ry9.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter_rn0ry9.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm3filter6_rn0ry9.pt', 'random_rec01_run1_alldeepconv_correctbyconvlstm2filter6_rn0ry9.pt'])
 
%% Cell type:code id:c384beaf tags:
 
``` python
relevant_models = list(pseudo_model_settings.keys())
relevant_models = thesis_filters
#relevant_models = ['allnoise_correctedscore', 'allnoise_correctbydeepconvfilter', 'allnoise_correctbyfcndaefilter', 'allnoise_correctbyfcndaefiltertest', 'allnoise_correctbyconvlstmfilter', 'allnoise_correctbyconvlstm2filter', 'allnoise_correctbyconvlstm3filter', 'alldeepconv_correctbyconvlstm3filter', 'alldeepconv_correctbyconvlstm3filter2', 'alldeepconv_correctbyconvlstm3filter3', 'alldeepconv_correctbyconvlstm3filter4','alldeepconv_correctbyconvlstm3filter5', 'alldeepconv_correctbyconvlstm3filter6']
# relevant_models = ['allnoise_correctedscore', 'allnoise_correctbydeepconvfilter', 'allnoise_correctbyfcndaefilter', 'allnoise_correctbyfcndaefiltertest', 'allnoise_correctbyconvlstmfilter', 'allnoise_correctbyconvlstm2filter', 'allnoise_correctbyconvlstm3filter', 'alldeepconv_correctbyconvlstm3filter', 'alldeepconv_correctbyconvlstm3filter2', 'alldeepconv_correctbyconvlstm3filter3']
# relevant_models = ['allnoise_correctbydeepconvfilter', 'allnoise_correctbyfcndaefiltertest', 'allnoise_correctbyfcndaefilter', 'alldeepconv_correctbyconvlstm3filter3']
#relevant_models = ['alldeepconv_correctbyconvlstm3filter6', ]
```
 
%% Cell type:code id:2c516d6b tags:
 
``` python
def get_target_pseudo_models(training_runs, training_manager):
target_models = []
collection = training_runs
for pseudo_model in relevant_models:
possible_models = get_models_with_generators(pseudo_model_settings[pseudo_model], training_manager.get_all_information(), collection)
for model in possible_models[:1]:
if model in training_runs and model not in target_models:
target_models.append(model)
return target_models
 
 
def calc_mean_values_of_pseudo_models(target_collections, target_value, relative_score=False):
mean_y_vals = []
mean_x_vals = []
mean_base = []
mean_base_inc = []
for target_collection in target_collections:
collection_config = evaluation_manager.get_collection_config(target_collection)
test_collection = evaluation_manager.get_test_sets_of_collection(target_collection)
target_models = evaluation_manager.get_run_of_collection(target_collection)
#print(target_models)
target_models = get_target_pseudo_models(target_models, evaluation_manager.get_training_manager_of_collection(target_collection))
# print(target_models)
model_evaluation = evaluation_manager.model_evaluation
base_model = collection_config['base_model']
base_inc_model = collection_config['base_inc_model']
pseudo_averages = calc_average_pseudo_model_evaluation(model_evaluation.predictions, target_value, target_models, test_collection)
base_averages = calc_average_pseudo_model_evaluation(model_evaluation.predictions, target_value, [base_model, base_inc_model], test_collection)
pseudo_average_values = []
for model_name in target_models:
pseudo_average_values.append(pseudo_averages[model_name])
 
if relative_score:
score_divider = base_averages[base_model]
base_averages[base_model] /= score_divider
base_averages[base_inc_model] /= score_divider
for pseudo_average in pseudo_averages:
pseudo_averages[pseudo_average] /= score_divider
 
# y_vals = list(pseudo_averages.values())
y_vals = [pseudo_averages[key] for key in target_models]
x_vals = remove_model_name_suffix(relevant_models)
if len(y_vals) == 1:
y_vals = np.repeat(y_vals, 2, axis=0)
x_vals = np.repeat(x_vals, 2, axis=0)
x_vals[-1] = '_'
mean_y_vals.append(y_vals)
mean_x_vals.append(x_vals)
mean_base.append(base_averages[base_model])
mean_base_inc.append(base_averages[base_inc_model])
 
mean_y_vals = np.array(mean_y_vals)
mean_y_vals = np.mean(mean_y_vals, axis=0)
mean_base = np.mean(np.array(mean_base))
mean_base_inc = np.mean(np.array(mean_base_inc))
x_vals = mean_x_vals[0]
x_vals_translated = [translate_setting(x_val) for x_val in x_vals]
 
return x_vals_translated, mean_y_vals, mean_base, mean_base_inc
 
def calc_mean_training_values_of_pseudo_models(target_collections, target_value, relative_score=False):
mean_y_vals = []
mean_x_vals = []
for target_collection in target_collections:
collection_config = evaluation_manager.get_collection_config(target_collection)
test_collection = evaluation_manager.get_test_sets_of_collection(target_collection)
target_models = evaluation_manager.get_run_of_collection(target_collection)
#print(target_models)
target_models = get_target_pseudo_models(target_models, evaluation_manager.get_training_manager_of_collection(target_collection))
# print(target_models)
training_manager = evaluation_manager.get_training_manager_of_collection(target_collection)
pseudo_average_values = []
training_sized = []
training_sizes_null = []
training_sizes_hw = []
for model_name in target_models:
model_info = training_manager.get_all_information()[model_name]
# print(model_info)
pseudo_average_values.append(model_info['pseudo_evaluation'][target_value])
training_sized.append(model_info['training_size']['overall_size'])
training_sizes_null.append(model_info['training_size']['null_size'])
training_sizes_hw.append(model_info['training_size']['hw_size'])
#pseudo_average_values.append(pseudo_averages[model_name])
 
if relative_score:
score_divider = base_averages[base_model]
for pseudo_average in pseudo_averages:
pseudo_averages[pseudo_average] /= score_divider
 
# y_vals = list(pseudo_averages.values())
y_vals = pseudo_average_values
x_vals = remove_model_name_suffix(relevant_models)
if len(y_vals) == 1:
y_vals = np.repeat(y_vals, 2, axis=0)
x_vals = np.repeat(x_vals, 2, axis=0)
x_vals[-1] = '_'
mean_y_vals.append(y_vals)
mean_x_vals.append(x_vals)
 
 
mean_y_vals = np.array(mean_y_vals)
mean_y_vals = np.mean(mean_y_vals, axis=0)
mean_training_sized = np.array(training_sized)
mean_training_sizes_null = np.array(training_sizes_null)
mean_training_sizes_hw = np.array(training_sizes_hw)
 
x_vals = mean_x_vals[0]
x_vals_translated = [translate_setting(x_val) for x_val in x_vals]
 
return x_vals_translated, mean_y_vals, mean_training_sized, mean_training_sizes_null, mean_training_sizes_hw
 
 
def calc_mean_confusion_matrix(target_collections):
mean_ppv = []
mean_npv = []
 
mean_tp = []
mean_fp = []
mean_tn = []
mean_fn = []
 
mean_training_sizes = []
mean_training_sizes_null = []
mean_training_sizes_hw = []
 
for target_collection in target_collections:
collection_config = evaluation_manager.get_collection_config(target_collection)
test_collection = evaluation_manager.get_test_sets_of_collection(target_collection)
target_models = evaluation_manager.get_run_of_collection(target_collection)
#print(target_models)
target_models = get_target_pseudo_models(target_models, evaluation_manager.get_training_manager_of_collection(target_collection))
# print(target_models)
training_manager = evaluation_manager.get_training_manager_of_collection(target_collection)
pseudo_average_values = []
ppv = []
npv = []
 
tp = []
fp = []
tn = []
fn = []
 
training_sizes = []
training_sizes_null = []
training_sizes_hw = []
for model_name in target_models:
model_info = training_manager.get_all_information()[model_name]
training_sizes.append(model_info['training_size']['overall_size'])
training_sizes_null.append(model_info['training_size']['null_size'])
training_sizes_hw.append(model_info['training_size']['hw_size'])
ppv.append(model_info['confusion_matrix']['ppv'])
npv.append(model_info['confusion_matrix']['npv'])
 
tp.append(model_info['confusion_matrix']['tp'])
fp.append(model_info['confusion_matrix']['fp'])
tn.append(model_info['confusion_matrix']['tn'])
fn.append(model_info['confusion_matrix']['fn'])
 
mean_ppv.append(ppv)
mean_npv.append(npv)
mean_training_sizes.append(training_sizes)
mean_training_sizes_null.append(training_sizes_null)
mean_training_sizes_hw.append(training_sizes_hw)
mean_tp.append(tp)
mean_fp.append(fp)
mean_tn.append(tn)
mean_fn.append(fn)
 
x_vals = remove_model_name_suffix(relevant_models)
x_vals_translated = [translate_setting(x_val) for x_val in x_vals]
 
mean_ppv = np.mean(np.array(mean_ppv), axis=0)
mean_npv = np.mean(np.array(mean_npv), axis=0)
 
mean_tp = np.mean(np.array(mean_tp), axis=0)
mean_fp = np.mean(np.array(mean_fp), axis=0)
mean_tn = np.mean(np.array(mean_tn), axis=0)
mean_fn = np.mean(np.array(mean_fn), axis=0)
 
 
mean_training_sizes = np.mean(np.array(mean_training_sizes), axis=0)
mean_training_sizes_null = np.mean(np.array(mean_training_sizes_null), axis=0)
mean_training_sizes_hw = np.mean(np.array(mean_training_sizes_hw), axis=0)
 
return x_vals_translated, mean_ppv, mean_npv, mean_tp, mean_fp, mean_tn, mean_fn, mean_training_sizes, mean_training_sizes_null, mean_training_sizes_hw
 
def plot_mean_average_of_pseudo_models(target_collections, target_value, relative_score=False):
x_vals_translated, mean_y_vals, mean_base, mean_base_inc = calc_mean_values_of_pseudo_models(target_collections, target_value, relative_score)
fig, ax = plt.subplots(figsize=(9,7))
ax.set_title(target_value)
ax.axhline(mean_base, linestyle=':', label='base', color='red')
ax.axhline(mean_base_inc, linestyle=':', label='inc', color='green')
ax.plot(x_vals_translated, mean_y_vals)
 
ax.set_xticklabels(x_vals_translated, rotation=75, ha='right', rotation_mode='anchor')
ax.grid(linestyle='-', linewidth=0.1)
ax.set_xlabel('modelname')
plt.tight_layout()
fig.show()
 
def barplot_mean_average_of_pseudo_models(target_collections, target_value, pseudo_target_value, relative_score=False, plot_pseudo=False, base_ax=None):
x_vals_translated, mean_y_vals, mean_base, mean_base_inc = calc_mean_values_of_pseudo_models(target_collections, target_value, relative_score)
pseudo_x_vals, pseudo_y_vals, mean_training_sized, mean_training_sizes_null, mean_training_sizes_hw = calc_mean_training_values_of_pseudo_models(target_collections, pseudo_target_value, relative_score)
# print(mean_training_sized)
 
if base_ax is None:
fig, ax = plt.subplots(figsize=(9,7))
# ax.set_title(target_value)
else:
ax = base_ax
 
barWidth = 0.55
if plot_pseudo:
barWidth = 0.35
r1 = np.arange(len(x_vals_translated))
r2 = [x + barWidth for x in r1]
r3 = [x + barWidth for x in r2]
 
bar_vals = []
for i in range(len(pseudo_y_vals)):
#bar_vals.append(pseudo_y_vals[i] * mean_training_sized[i])
bar_vals.append(pseudo_y_vals[i] * mean_training_sizes_null[i])
#bar_vals.append(pseudo_y_vals[i] * min(1.0, mean_training_sizes_hw[i]))
#bar_vals.append(pseudo_y_vals[i])
 
p1 = ax.bar(r1, mean_y_vals, width=barWidth, edgecolor='white', label='model evaluation')
# ax.bar(r1, mean_y_vals*mean_training_sized, width=barWidth,edgecolor='black', color='cornflowerblue', alpha=0.5, label='scaled model evaluation')
if plot_pseudo:
p2 = ax.bar(r2, bar_vals, width=barWidth, edgecolor='white', label='train set evaluation')
#bars = ax.bar(r3, mean_training_sized, width=barWidth, edgecolor='white')
 
 
#for bars in ax.containers:
# #ax.bar_label(bars)
# texts = ax.bar_label(bars, rotation=90, label_type='center', backgroundcolor=(1, 1, 1, 0.3), fmt='%.3f')
# #print(texts)
# for text in texts:
# bb = text.get_bbox_patch()
# bb.set_boxstyle("square", pad=0)
# text.set(y=15)
 
ax.bar_label(p1,rotation=90, label_type='center', backgroundcolor=(1, 1, 1, 0.3), fmt='%.3f')
 
ax.axhline(mean_base, linestyle=':', label='base model', color='red')
ax.axhline(mean_base_inc, linestyle=':', label='base inc model', color='green')
ax.set_xticklabels(x_vals_translated, rotation=75, ha='right', rotation_mode='anchor')
ax.grid(linestyle='-', linewidth=0.1)
ax.set_xlabel('configuration', fontweight='bold')
 
# plt.xlabel('configuration', fontweight='bold')
tick_shift = barWidth
if plot_pseudo:
tick_shift = (barWidth/2)
ax.set_xticks([r + barWidth - tick_shift for r in range(len(x_vals_translated))], x_vals_translated, rotation=75, ha='right', rotation_mode='anchor')
# plt.xticks(x_vals_translated, rotation=75, ha='right', rotation_mode='anchor')
translate_plot_value = {'specificity': 'specificity', 'sensitivity': 'sensitivity', 'f1': 'F1 score', 's1': 'S score', 'mcc': 'MCC score'}
ax.set_ylabel(translate_plot_value[target_value])
# plt.tight_layout()
#plt.subplots_adjust(top=0.9, bottom=0.28)
ax.legend(loc='lower right' )
if base_ax is None:
fig.legend()
fig.show()
 
 
def barplot_mean_training_data(target_collections, target_values, base_ax=None):
mean_training_evaluations = []
for target_value in target_values:
mean_training_evaluations.append(calc_mean_training_values_of_pseudo_models(target_collections, target_value, False))
if base_ax is None:
fig, ax = plt.subplots(figsize=(9,5))
# ax.set_title(target_value)
else:
ax = base_ax
 
barWidth = 0.8 / len(target_values)
r = [np.arange(len(mean_training_evaluations[0][0]))]
for i in range(1, len(target_values)):
r.append([x + barWidth for x in r[i-1]])
 
translate_plot_value = {'specificity': 'specificity', 'sensitivity': 'sensitivity', 'f1': 'F1 score', 's1': 'S score', 'mcc': 'MCC score'}
color_pairs = (('tab:blue', 'blue'), ('tab:orange', 'orange'), ('tab:green', 'green'), ('tab:red', 'red'))
for i in range(len(target_values)):
p1 = ax.bar(r[i], mean_training_evaluations[i][1], width=barWidth, color=color_pairs[i][0], edgecolor=None, label=translate_plot_value[target_values[i]])
 
scaled_vals = mean_training_evaluations[i][1]*mean_training_evaluations[i][2]
dividers = scaled_vals
dividers[scaled_vals >= mean_training_evaluations[i][1] - 0.01] = -1
ax.bar(r[i], dividers, width=barWidth, color='white', alpha=0.4, edgecolor=None)
ax.bar(r[i], 0.005, bottom=scaled_vals, color='black', width=barWidth)
 
tick_shift = barWidth/2
ax.set_xticks([r + barWidth + tick_shift for r in range(len(mean_training_evaluations[0][0]))], mean_training_evaluations[0][0], rotation=55, ha='right', rotation_mode='anchor')
 
 
ax.set_xlabel('configuration', fontweight='bold')
ax.legend(loc='lower right' )
ax.grid(linestyle='-', linewidth=0.1)
ax.set_ylim(0,1.1)
if base_ax is None:
fig.tight_layout()
fig.show()
 
 
def barplot_confusion_matrix_training_data(target_collections, base_ax=None):
x_vals_translated, mean_ppv, mean_npv, mean_tp, mean_fp, mean_tn, mean_fn, mean_training_sizes, mean_training_sizes_null, mean_training_sizes_hw = calc_mean_confusion_matrix(target_collections)
if base_ax is None:
fig, ax = plt.subplots(figsize=(9,5))
# ax.set_title(target_value)
else:
ax = base_ax
 
display_bars = 4
 
 
barWidth = 0.8 / display_bars
r = [np.arange(len(x_vals_translated))]
for i in range(1, display_bars):
r.append([x + barWidth for x in r[i-1]])
 
translate_plot_value = {'specificity': 'specificity', 'sensitivity': 'sensitivity', 'f1': 'F1 score', 's1': 'S score', 'mcc': 'MCC score'}
 
#p1 = ax.bar(r[0], mean_ppv[:, 1], width=barWidth, edgecolor='white', label='mean_ppv')
#p2 = ax.bar(r[1], mean_npv[:, 1], width=barWidth, edgecolor='white', label='mean_npv')
p3 = ax.bar(r[0], mean_tp, width=barWidth, color='tab:green', label='tp')
p4 = ax.bar(r[1], mean_fp, width=barWidth, color='tab:red', label='fp')
p5 = ax.bar(r[2], mean_tn, width=barWidth, color='tab:blue', label='tn')
p5 = ax.bar(r[3], mean_fn, width=barWidth, color='tab:orange', label='fn')
#p6 = ax.bar(r[4], mean_training_sizes, width=barWidth, color='tab:olive', label='training_size')
 
 
tick_shift = barWidth/2
ax.set_xticks([r + barWidth + tick_shift for r in range(len(x_vals_translated))], x_vals_translated, rotation=55, ha='right', rotation_mode='anchor')
ax.set_xlabel('configuration', fontweight='bold')
#ax.set_ylim(0, 1.4)
ax.legend(loc='lower right' )
ax.grid(linestyle='-', linewidth=0.1)
if base_ax is None:
#fig.legend()
fig.tight_layout()
fig.show()
 
table_dict = dict()
for i in range(len(x_vals_translated)):
table_dict[x_vals_translated[i]] = {'tp': mean_tp[i], 'fp': mean_fp[i], 'tn': mean_tn[i], 'fn': mean_fn[i]}
 
return table_dict
 
```
 
%% Cell type:code id:6ca0d24a tags:
 
``` python
include_collections = []
include_collections += ['synthetic_01', 'synthetic_02', 'synthetic_10']
include_collections += ['recorded_01', 'recorded_02']
#include_collections = ['recorded_01', 'recorded_02']
 
include_collections = ['recorded_01']
 
fig, axes = plt.subplots(1,2, figsize=(15, 6))
barplot_mean_average_of_pseudo_models(include_collections, 'specificity', 'specificity', plot_pseudo=False, base_ax=axes[0])
barplot_mean_average_of_pseudo_models(include_collections, 'sensitivity', 'sensitivity', plot_pseudo=False, base_ax=axes[1])
fig.tight_layout()
#fig.legend()
fig.show()
 
fig, axes = plt.subplots(1,2, figsize=(15, 6))
barplot_mean_average_of_pseudo_models(include_collections, 'f1', 'f1', plot_pseudo=False, base_ax=axes[0])
barplot_mean_average_of_pseudo_models(include_collections, 's1', 's1', plot_pseudo=False, base_ax=axes[1])
fig.tight_layout()
#fig.legend()
fig.show()
```
 
%% Output
 
 
 
<ipython-input-27-30713955afc6>:246: UserWarning: FixedFormatter should only be used together with FixedLocator
ax.set_xticklabels(x_vals_translated, rotation=75, ha='right', rotation_mode='anchor')
 
 
 
%% Cell type:code id:246d8e98 tags:
 
``` python
include_collections = []
include_collections += ['synthetic_01', 'synthetic_02', 'synthetic_10']
include_collections += ['recorded_01', 'recorded_02']
#include_collections = ['recorded_01', 'recorded_02']
 
 
barplot_mean_training_data(include_collections, ['specificity', 'sensitivity', 'f1', 's1'])
table_dict = barplot_confusion_matrix_training_data(include_collections)
data_frame = pd.DataFrame.from_dict(table_dict, orient='index')
data_frame
```
 
%% Output
 
 
 
 
 
tp fp tn fn
all 0.782806 0.109959 0.890041 0.217194
high_conf 0.335298 0.004089 0.623328 0.009102
scope 0.754095 0.031934 0.047074 0.208741
all_corrected_null 0.782806 0.081051 0.918949 0.217194
scope_corrected_null 0.754095 0.003026 0.075982 0.208741
all_corrected_null_hwgt 0.782806 0.081063 0.918937 0.217194
scope_corrected_null_hwgt 0.754095 0.003038 0.075969 0.208741
all_null_hwgt 0.994162 0.000000 1.000000 0.005838
all_null_score 0.768328 0.000409 0.999591 0.231672
all_null_deepconv 0.767056 0.000684 0.999316 0.232944
all_null_fcndae 0.868311 0.001502 0.998498 0.131689
all_null_convlstm1 0.872169 0.001494 0.998506 0.127831
all_null_convlstm2 0.870221 0.001429 0.998571 0.129779
all_null_convlstm3 0.874198 0.001565 0.998435 0.125802
all_cnn_convlstm3 0.878977 0.004642 0.917894 0.120518
all_cnn_convlstm2_hard 0.838896 0.001429 0.917125 0.129274
all_cnn_convlstm3_hard 0.853719 0.001505 0.917048 0.114451
 
%% Cell type:markdown id:028749d9 tags:
 
# Randomized
 
%% Cell type:code id:d3d3e7a0 tags:
 
``` python
def get_multiple_target_pseudo_models(training_runs, training_manager):
target_models = dict()
collection = training_runs
for pseudo_model in relevant_models:
possible_models = get_models_with_generators(pseudo_model_settings[pseudo_model], training_manager.get_all_information(), collection)
for model in possible_models:
if model in training_runs and model not in target_models:
if pseudo_model not in target_models:
target_models[pseudo_model] = []
target_models[pseudo_model].append(model)
return target_models
 
def get_models_by_filter(target_filter, relevant_model_info):
models = []
for model, values in relevant_model_info.items():
add_model = True
#print(model, values)
for target_key, target_value in target_filter.items():
#print(target_key, target_value, values)
if values[target_key] != target_value:
add_model = False
break
if add_model:
models.append(model)
return models
 
def get_model_infos(models, all_info):
infos = dict()
for model in models:
model_info = all_info[model]
evaluation = model_info['pseudo_evaluation']
training_size = model_info['training_size']
confusion_matrix = model_info['confusion_matrix']
info_entry = {'s1': evaluation['s1'], 'overall_size': training_size['overall_size'], 'null_size': training_size['null_size'], 'hw_size': training_size['hw_size'],
'tp': confusion_matrix['tp'], 'fp': confusion_matrix['fp'], 'tn': confusion_matrix['tn'], 'fn': confusion_matrix['fn'],
'random_no': model_info['random_no'], 'random_yes': model_info['random_yes'], 'hw_regions': training_size['hw_regions']}
infos[model] = info_entry
return infos
 
# target_collections, target_value, pseudo_target_value, relative_score=False, plot_pseudo=False, base_ax=None
def plot_randomized_average(target_collections, target_score, target_random_value, base_ax=None):
fig, ax = plt.subplots(figsize=(10,5))
 
ax.set_title(target_score)
ax.set_ylabel(target_score)
ax.set_xlabel('evaluation reliability')
target_randoms = []
for i in np.arange(0.2, 1.05, 0.1):
i = np.around(i, decimals=3)
target_randoms.append(i)
target_steady_value = 'random_no' if target_random_value == 'random_yes' else 'random_yes'
 
overall_info = dict()
relevant_models = ['alldeepconv_correctbyconvlstm3filter6']
for pseudo_model_name in relevant_models:
mean_y = []
for target_collection in target_collections:
# print(target_collection)
collection_config = evaluation_manager.get_collection_config(target_collection)
#print(collection_config)
test_collection = evaluation_manager.get_test_sets_of_collection(target_collection)
target_models = evaluation_manager.get_run_of_collection(target_collection)
#print(target_models)
target_models = get_multiple_target_pseudo_models(target_models, evaluation_manager.get_training_manager_of_collection(target_collection))
#print(target_models)
training_manager = evaluation_manager.get_training_manager_of_collection(target_collection)
 
filtered_target_models = []
for random_value in target_randoms:
model_filter = {target_random_value: random_value, target_steady_value: 1.00}
possible_models = get_models_by_filter(model_filter, training_manager.get_information_of_models(target_models[pseudo_model_name]))
target_model = None
for pseudo_model_list in target_models[pseudo_model_name]:
for possible_model in possible_models:
if possible_model in pseudo_model_list:
# print(possible_model, pseudo_model_list)
target_model = possible_model
break
#print(target_model)
filtered_target_models.append(target_model)
 
pseudo_averages = calc_average_pseudo_model_evaluation(evaluation_manager.model_evaluation.predictions, target_score, filtered_target_models, test_collection)
pseudo_average_values = []
for model_name in filtered_target_models:
pseudo_average_values.append(pseudo_averages[model_name])
infos = get_model_infos(filtered_target_models, training_manager.get_all_information())
overall_info.update(infos)
y_vals = [pseudo_averages[key] for key in filtered_target_models]
mean_y.append(y_vals)
mean_y = np.mean(np.array(mean_y), axis=0)
ax.plot(target_randoms, mean_y, label=translate_setting(pseudo_model_name))
 
 
# plt.xticks(rotation=45, ha='right')
plt.subplots_adjust(left=0.1, right=0.7, top=0.9, bottom=0.15)
fig.legend()
 
return ax, overall_info
 
 
include_collections = ['random_synthetic_01']
_, info = plot_randomized_average(include_collections, 's1', 'random_no')
data_frame = pd.DataFrame.from_dict(info, orient='index')
data_frame.filter(like='alldeepconv', axis=0)
```
 
%% Output
 
 
 
s1 overall_size \
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.954194 0.859584
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.957823 0.850537
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.955018 0.848911
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.954794 0.847177
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.956105 0.847177
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.955401 0.847177
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.954685 0.847177
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.954470 0.847177
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.831699 0.836112
null_size hw_size \
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.858235 0.963931
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.848973 0.972416
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.847281 0.976446
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.845640 0.965578
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.845526 0.975093
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.845649 0.965739
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.845623 0.966901
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.845540 0.976769
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.841473 0.328709
tp fp \
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.923913 0.000442
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.930751 0.000462
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.927913 0.000541
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.927913 0.000423
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.930751 0.000504
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.925236 0.000456
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.928913 0.000425
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.925913 0.000553
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.327709 0.000015
tn fn \
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.857549 0.076087
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.848379 0.069249
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.846654 0.072087
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.845013 0.072087
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.844932 0.069249
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.844980 0.074764
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.845011 0.071087
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.844883 0.074087
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.840298 0.116056
random_no random_yes \
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.2 1.0
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.3 1.0
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.4 1.0
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.5 1.0
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.6 1.0
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.7 1.0
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.8 1.0
random_01_run1_alldeepconv_correctbyconvlstm3fi... 0.9 1.0
random_01_run1_alldeepconv_correctbyconvlstm3fi... 1.0 1.0
hw_regions
random_01_run1_alldeepconv_correctbyconvlstm3fi... 26
random_01_run1_alldeepconv_correctbyconvlstm3fi... 26
random_01_run1_alldeepconv_correctbyconvlstm3fi... 26
random_01_run1_alldeepconv_correctbyconvlstm3fi... 26
random_01_run1_alldeepconv_correctbyconvlstm3fi... 26
random_01_run1_alldeepconv_correctbyconvlstm3fi... 26
random_01_run1_alldeepconv_correctbyconvlstm3fi... 26
random_01_run1_alldeepconv_correctbyconvlstm3fi... 26
random_01_run1_alldeepconv_correctbyconvlstm3fi... 6
 
%% Cell type:markdown id:07cd0892 tags:
 
# Old stuff
 
%% Cell type:code id:e562d499 tags:
 
``` python
# Synthetic
predictions_db = './data/cluster/pseudo_collections/synthetic_predictions_db'
dataset_db = './data/synthetic_dataset_db'
training_db = './data/cluster/pseudo_collections/synthetic_pseudo_training_db'
models_directory = './data/cluster/pseudo_collections/synthetic_models/'
base_models = ['HandWashingDeepConvLSTMA_trunc_01.pt', 'HandWashingDeepConvLSTMA_trunc_02.pt','HandWashingDeepConvLSTMA_trunc_10.pt', 'HandWashingDeepConvLSTMA_trunc_01.pt']
base_inc_models = ['synthetic_inc_01.pt', 'synthetic_inc_02.pt', 'synthetic_inc_10.pt']
test_collection_names = ['01_test', '02_test', '10_test']
training_collection_names = ['01_training', '02_training', '10_training']
collection_training_run_assignments = [('01_test', 'pseudo_collection_01_run1'), ('02_test', 'pseudo_collection_02_run1'), ('10_test', 'pseudo_collection_10_run1')]
```
 
%% Cell type:code id:f44c7fa8 tags:
 
``` python
# Synthetic randomized
predictions_db = './data/cluster/pseudo_randomized/synthetic_predictions_db'
dataset_db = './data/synthetic_dataset_db'
training_db = './data/cluster/pseudo_randomized/pseudo_training_db'
models_directory = './data/cluster/pseudo_randomized/pseudo_models/'
base_models = ['HandWashingDeepConvLSTMA_trunc_01.pt']
base_inc_models = ['synthetic_inc_01.pt']
test_collection_names = ['01_test']
training_collection_names = ['01_training']
collection_training_run_assignments = [('01_test', 'pseudo_randomeval_01_run1')]
```
 
%% Cell type:code id:1cba0e77 tags:
 
``` python
# Recordings
predictions_db = './data/cluster/pseudo_collections/recorded_predictions_db'
dataset_db = './data/recorded_dataset_db'
training_db = './data/cluster/pseudo_collections/recorded_pseudo_training_db'
models_directory = './data/cluster/pseudo_collections/recorded_models/'
base_models = ['DeepConvLSTMA_statedict.pt', 'DeepConvLSTMA_statedict.pt']
base_inc_models = ['recorded_inc_01.pt', 'recorded_inc_02.pt']
test_collection_names = ['01_test', '02_test']
training_collection_names = ['01_training', '02_training']
collection_training_run_assignments = [('01_test', 'pseudo_collection_01_run1'), ('02_test', 'pseudo_collection_02_run1')]
```
 
%% Cell type:code id:59d9a9a7 tags:
 
``` python
# Recordings
predictions_db = './data/cluster/recorded_pseudo_predictions_db_2'
dataset_db = './data/recorded_dataset_db_2'
training_db = './data/cluster/recorded_pseudo_training_db_2'
models_directory = './data/cluster/recorded_pseudo_models_2/'
base_models = ['DeepConvLSTMA_statedict.pt',]
base_inc_models = ['recorded_inc_01.pt',]
test_collection_names = ['01_test', ]
training_collection_names = ['01_training',]
collection_training_run_assignments = [('01_test', 'pseudo_collection_01_run1'), ]
```
 
%% Cell type:code id:5ee876c5 tags:
 
``` python
model_evaluation = ModelEvaluation()
model_evaluation.load_predictions(predictions_db)
 
dataset_manager = DatasetManager(dataset_db)
 
unseen_datasets = []
test_collections = dict()
for test_collection_name in test_collection_names:
unseen_datasets.append(list(dataset_manager.filter_by_category(test_collection_name).keys()))
test_collections[test_collection_name] = list(dataset_manager.filter_by_category(test_collection_name).values())
 
training_collections = []
for training_collection_name in training_collection_names:
training_collections.append(dataset_manager.filter_by_category(training_collection_name))
 
model_evaluation.clear_datasets()
for test_collection in test_collections.values():
model_evaluation.add_collection(test_collection)
 
training_manager = TrainingsManager(training_db)
training_runs = training_manager.get_all_training_runs()
print(training_runs.keys())
 
for run in training_runs.keys():
print('add run:', run)
for model in training_runs[run]:
model_evaluation.add_model(models_directory + model)
 
for base_model in base_models:
model_evaluation.add_model('./data/' + base_model)
for inc_model in base_inc_models:
model_evaluation.add_model('./data/' + inc_model)
 
#for trained_collection in training_collections:
# model_evaluation.predict_models_on_collection(base_models, trained_collection.values())
 
for collection_training_run_assignment in collection_training_run_assignments:
test_collection = test_collections[collection_training_run_assignment[0]]
models_list = training_runs[collection_training_run_assignment[1]]
#print('assign', test_collection, 'to', models_list)
model_evaluation.assign_collection_to_model_list(test_collection, models_list)
```
 
%% Cell type:code id:cb82065e tags:
 
``` python
# model_evaluation.clear_evaluations_of_model('pseudo_collection_01_run1_allnoise_correctbyconvlstmfilter.pt')
# model_evaluation.clear_evaluations_of_model('pseudo_collection_01_run1_allnoise_correctbyconvlstm3filter.pt')
#model_evaluation.clear_evaluations_of_model('pseudo_collection_01_run1_alldeepconv_correctbyconvlstm3filter.pt')
#model_evaluation.clear_evaluations_of_model('pseudo_collection_01_run1_alldeepconv_correctbyconvlstm3filter2.pt')
#model_evaluation.clear_evaluations_of_model('pseudo_collection_01_run1_alldeepconv_correctbyconvlstm3filter3.pt')
 
#model_evaluation.clear_evaluations_of_model('pseudo_collection_01_run1_allnoise_correctbyconvlstm3filter.pt')
#model_evaluation.clear_evaluations_of_model('pseudo_collection_01_run1_allnoise_correctbyconvlstm2filter.pt')
#model_evaluation.clear_evaluations_of_model('pseudo_collection_01_run1_allnoise_correctbyconvlstmfilter.pt')
#model_evaluation.clear_evaluations_of_model('pseudo_collection_01_run1_allnoise_correctbyfcndaefiltertest.pt')
#model_evaluation.clear_evaluations_of_model('pseudo_collection_01_run1_allnoise_correctbyfcndaefilter.pt')
#model_evaluation.clear_evaluations_of_model('pseudo_collection_01_run1_allnoise_correctbydeepconvfilter.pt')
model_evaluation.clear_evaluations_of_model('pseudo_collection_10_run1_allnoise_correctedscore.pt')
```
 
%% Cell type:code id:ab8b166e tags:
 
``` python
model_evaluation.clear_evaluations()
```
 
%% Cell type:code id:92ee8a8a tags:
 
``` python
model_evaluation.do_predictions(use_soft=False)
```
 
%% Cell type:code id:3398d867 tags:
 
``` python
model_evaluation.save_predictions(predictions_db)
```
 
%% Cell type:code id:8bac064e tags:
 
``` python
evaluations = model_evaluation.get_evaluations(calc_averages=True, sort_by='model', include_datasets=[dataset.name for dataset in test_collections['01_test']])
 
evaluations
```
 
%% Cell type:code id:85417e82 tags:
 
``` python
pseudo_randomeval_01_run1_allnoise_correctedhwgt_rn0ry2.pt
#evaluations = model_evaluation.get_evaluations(include_models=training_runs['pseudo_collection_01_run1'], calc_averages=True, sort_by='S1')
#evaluations = model_evaluation.get_evaluations(include_models=training_runs['pseudo_collection_01_run1'])
#evaluations = model_evaluation.get_evaluations(include_datasets=[dataset.name for dataset in test_collections['10_test']], calc_averages=False, sort_by='f1')
evaluations = model_evaluation.get_evaluations(calc_averages=True, sort_by='f1', include_datasets=[dataset.name for dataset in test_collections['01_test']])
 
evaluations
```
 
%% Cell type:code id:544f21ef tags:
 
``` python
requests = 0
participant = 1
 
observed_collection = list(dataset_manager.filter_by_category(training_collection_names[participant]).values())
for dataset in observed_collection:
evaluations = dataset.get_indicators()[1]
for evaluation in evaluations:
if evaluation[1] != Indicators.NEUTRAL:
requests += 1
print('Num requests:', requests)
print('requests per dataset:', requests/len(observed_collection))
```
 
%% Cell type:code id:12e7b48a tags:
 
``` python
relevant_models = list(pseudo_model_settings.keys())
relevant_models = thesis_filters
#relevant_models = ['allnoise_correctedscore', 'allnoise_correctbydeepconvfilter', 'allnoise_correctbyfcndaefilter', 'allnoise_correctbyfcndaefiltertest', 'allnoise_correctbyconvlstmfilter', 'allnoise_correctbyconvlstm2filter', 'allnoise_correctbyconvlstm3filter', 'alldeepconv_correctbyconvlstm3filter', 'alldeepconv_correctbyconvlstm3filter2', 'alldeepconv_correctbyconvlstm3filter3', 'alldeepconv_correctbyconvlstm3filter4','alldeepconv_correctbyconvlstm3filter5', 'alldeepconv_correctbyconvlstm3filter6']
# relevant_models = ['allnoise_correctedscore', 'allnoise_correctbydeepconvfilter', 'allnoise_correctbyfcndaefilter', 'allnoise_correctbyfcndaefiltertest', 'allnoise_correctbyconvlstmfilter', 'allnoise_correctbyconvlstm2filter', 'allnoise_correctbyconvlstm3filter', 'alldeepconv_correctbyconvlstm3filter', 'alldeepconv_correctbyconvlstm3filter2', 'alldeepconv_correctbyconvlstm3filter3']
# relevant_models = ['allnoise_correctbydeepconvfilter', 'allnoise_correctbyfcndaefiltertest', 'allnoise_correctbyfcndaefilter', 'alldeepconv_correctbyconvlstm3filter3']
#relevant_models = ['alldeepconv_correctbyconvlstm3filter6', ]
```
 
%% Cell type:code id:4376c450 tags:
 
``` python
def plot_collection(target_value, collection_name, test_collection, base_model, base_inc_model, ax=None, plot_baelines=True):
if ax is None:
fig, ax = plt.subplots()
ax.set_title(target_value)
collection = training_runs[collection_name]
for i, target_dataset in enumerate(test_collection):
y_vals = []
for pseudo_model_name in relevant_models:
generators = pseudo_model_settings[pseudo_model_name]
models = get_models_with_generators(generators, training_manager.get_all_information(), collection)
for model in models[:1]:
if model in training_runs[collection_name]:
evaluation = (model, target_dataset)
val = getattr(model_evaluation.predictions[evaluation].evaluation, target_value)
y_vals.append(val)
 
y_vals = np.array(y_vals)
x_vals = np.array(relevant_models)
if y_vals.shape[0] == 1:
y_vals = np.repeat(y_vals, 2, axis=0)
x_vals = np.repeat(x_vals, 2, axis=0)
x_vals[-1] = '_'
label = None
if plot_baelines:
label = target_dataset
print(x_vals, y_vals.shape[0])
base_line, = ax.plot(x_vals, y_vals, label=label)
 
label = None
if plot_baelines and i == 0:
label = 'genereal model'
val = getattr(model_evaluation.predictions[(base_model, target_dataset)].evaluation, target_value)
ax.axhline(val, linestyle='--', label=label, color=base_line.get_color())
if plot_baelines and i == 0:
label = 'gt inc model'
val = getattr(model_evaluation.predictions[(base_inc_model, target_dataset)].evaluation, target_value)
ax.axhline(val, linestyle=':', label=label, color=base_line.get_color())
 
 
#y_vals = []
#for average in pseudo_evaluations_avg:
# y_vals.append(average[1][target_value])
#ax.plot(list(pseudo_model_settings.keys()), y_vals, label='Pseudo average hard')
 
#y_vals = []
#for average in pseudo_evaluations_avg:
# y_vals.append(average[2][target_value])
#ax.plot(list(pseudo_model_settings.keys()), y_vals, label='Pseudo average soft')
 
 
ax.tick_params(labelrotation=75)
ax.grid(linestyle='-', linewidth=0.1)
 
#val = training_manager.get_information(model)['pseudo_evaluation'][target_value]
 
def plot_pseudo_evaluation(target_value, collection_name, ax):
y_vals = []
collection = training_runs[collection_name]
for pseudo_model_name, generators in pseudo_model_settings.items():
if relevant_models is not None and pseudo_model_name not in relevant_models:
continue
models = get_models_with_generators(generators, training_manager.get_all_information(), collection)
for model in models[:1]:
if model in training_runs[collection_name]:
val = training_manager.get_information(model)['pseudo_evaluation'][target_value]
y_vals.append(val)
y_vals = np.array(y_vals)
if relevant_models is None:
ax.plot(list(pseudo_model_settings.keys()), y_vals, label='label evaluation', linestyle='-.')
else:
ax.plot(relevant_models, y_vals, label='label evaluation', linestyle='-.')
 
def plot_all_for_collection(collection_name, test_collection, base_model, base_inc_model):
fig, axes = plt.subplots(4, figsize=(10,32))
#fig.tight_layout()
plot_collection('specificity', collection_name, test_collection, base_model, base_inc_model, axes[0], True)
plot_pseudo_evaluation('specificity', collection_name, axes[0])
plot_collection('sensitivity', collection_name, test_collection, base_model, base_inc_model, axes[1], False)
plot_pseudo_evaluation('sensitivity', collection_name, axes[1])
plot_collection('f1', collection_name, test_collection, base_model, base_inc_model, axes[2], False)
plot_pseudo_evaluation('f1', collection_name, axes[2])
plot_collection('s1', collection_name, test_collection, base_model, base_inc_model, axes[3], False)
plot_pseudo_evaluation('s1', collection_name, axes[3])
 
plt.subplots_adjust(hspace=0.8)
plt.subplots_adjust(bottom=0.0)
fig.legend()
fig.show()
 
```
 
%% Cell type:code id:54139635 tags:
 
``` python
# print(pseudo_model_settings.keys())
target_participant = 0
model_collection = collection_training_run_assignments[target_participant][1]
test_datasets = [dataset.name for dataset in test_collections[test_collection_names[target_participant]]]
base_model = base_models[target_participant]
base_inc_model = base_inc_models[target_participant]
plot_all_for_collection(model_collection, test_datasets, base_model, base_inc_model)
```
 
%% Cell type:code id:a2230ff7 tags:
 
``` python
def get_target_pseudo_models(collection_name):
target_models = []
collection = training_runs[collection_name]
for pseudo_model in relevant_models:
possible_models = get_models_with_generators(pseudo_model_settings[pseudo_model], training_manager.get_all_information(), collection)
for model in possible_models[:1]:
if model in training_runs[collection_name] and model not in target_models:
target_models.append(model)
return target_models
 
def plot_average_of_pseudo_models(collection_name, target_value, test_dataset, base_model, base_inc_model, relative_score=False):
target_models = get_target_pseudo_models(collection_name)
pseudo_averages = calc_average_pseudo_model_evaluation(model_evaluation.predictions, target_value, target_models, test_dataset)
base_averages = calc_average_pseudo_model_evaluation(model_evaluation.predictions, target_value, [base_model, base_inc_model], test_dataset)
pseudo_average_values = []
for model_name in target_models:
pseudo_average_values.append(pseudo_averages[model_name])
 
 
 
if relative_score:
score_divider = base_averages[base_model]
base_averages[base_model] /= score_divider
base_averages[base_inc_model] /= score_divider
for pseudo_average in pseudo_averages:
pseudo_averages[pseudo_average] /= score_divider
 
y_vals = list(pseudo_averages.values())
x_vals = remove_model_name_suffix(relevant_models)
if len(y_vals) == 1:
y_vals = np.repeat(y_vals, 2, axis=0)
x_vals = np.repeat(x_vals, 2, axis=0)
x_vals[-1] = '_'
 
 
fig, ax = plt.subplots(figsize=(9,7))
ax.set_title(target_value)
ax.axhline(base_averages[base_model], linestyle=':', label='base', color='red')
ax.axhline(base_averages[base_inc_model], linestyle=':', label='inc', color='green')
ax.plot(x_vals, y_vals)
 
ax.set_xticklabels(x_vals, rotation=75, ha='right', rotation_mode='anchor')
ax.grid(linestyle='-', linewidth=0.1)
ax.set_xlabel('modelname')
plt.tight_layout()
fig.show()
 
 
target_participant = 0
test_datasets = [dataset.name for dataset in test_collections[test_collection_names[target_participant]]]
base_model = base_models[target_participant]
base_inc_model = base_inc_models[target_participant]
 
run_name = collection_training_run_assignments[target_participant][1]
plot_average_of_pseudo_models(run_name, 'specificity', test_datasets, base_model, base_inc_model)
plot_average_of_pseudo_models(run_name, 'sensitivity', test_datasets, base_model, base_inc_model)
plot_average_of_pseudo_models(run_name, 'f1', test_datasets, base_model, base_inc_model, relative_score=False)
plot_average_of_pseudo_models(run_name, 's1', test_datasets, base_model, base_inc_model)
plot_average_of_pseudo_models(run_name, 'mcc', test_datasets, base_model, base_inc_model)
```
 
%% Cell type:code id:26b45e5e tags:
 
``` python
def get_models_by_filter(target_filter, relevant_model_info):
models = []
for model, values in relevant_model_info.items():
add_model = True
for target_key, target_value in target_filter.items():
#print(target_key, target_value, values)
if values[target_key] != target_value:
add_model = False
break
if add_model:
models.append(model)
return models
 
def get_model_by_filter(target_filter, relevant_model_info):
for model, values in relevant_model_info.items():
add_model = True
for target_key, target_value in target_filter.items():
if values[target_key] != target_value:
add_model = False
break
 
if add_model:
return model
 
 
def plot_randomized(target_dataset, target_score, target_values, model_infos, collection_name):
fig, ax = plt.subplots(figsize=(9,5))
 
ax.set_title(target_score)
ax.set_ylabel(target_score)
ax.set_xlabel('evaluation reliability')
 
 
 
collection = training_runs[collection_name]
pseudo_models = dict()
for pseudo_model_name in relevant_models:
if pseudo_model_name not in pseudo_models:
pseudo_models[pseudo_model_name] = []
pseudo_models[pseudo_model_name].append(get_models_with_generators(pseudo_model_settings[pseudo_model_name], training_manager.get_all_information(), collection))
 
for pseudo_model_name in relevant_models:
y_vals = []
x_vals = []
for target_value in target_values:
model_filter = {'random_no': target_value[0], 'random_yes': target_value[1]}
possible_models = get_models_by_filter(model_filter, model_infos)
target_model = None
for pseudo_model_list in pseudo_models[pseudo_model_name]:
for possible_model in possible_models:
if possible_model in pseudo_model_list:
#print(possible_model, pseudo_model_list)
target_model = possible_model
break
#print(target_model)
if target_model is not None:
evaluation = (target_model, target_dataset)
val = getattr(model_evaluation.predictions[evaluation].evaluation, target_score)
y_vals.append(val)
nonzero_val = target_value[0]
if nonzero_val == 0 or nonzero_val == 1:
nonzero_val = target_value[1]
x_vals.append(nonzero_val)
 
ax.plot(x_vals, y_vals, label=pseudo_model_name)
 
#val = getattr(model_evaluation.predictions[('HandWashingDeepConvLSTMA_trunc_01.pt', target_dataset)].evaluation, target_score)
#ax.axhline(val, color='red', linestyle='--', label='genereal model')
#val = getattr(model_evaluation.predictions[('synthetic_inc_01.pt', target_dataset)].evaluation, target_score)
#ax.axhline(val, color='green', linestyle='--', label='gt inc model')
 
 
 
# plt.xticks(rotation=45, ha='right')
try:
fig.legend(loc='lower right')
except NameError:
pass
 
return ax
 
 
def plot_randomized_average(test_collection_name, target_score, target_values, model_infos, collection_name):
fig, ax = plt.subplots(figsize=(9,5))
 
ax.set_title(target_score)
ax.set_ylabel(target_score)
ax.set_xlabel('evaluation reliability')
 
 
 
collection = training_runs[collection_name]
pseudo_models = dict()
for pseudo_model_name in relevant_models:
if pseudo_model_name not in pseudo_models:
pseudo_models[pseudo_model_name] = []
pseudo_models[pseudo_model_name].append(get_models_with_generators(pseudo_model_settings[pseudo_model_name], training_manager.get_all_information(), collection))
 
for pseudo_model_name in relevant_models:
y_vals = []
x_vals = []
for target_value in target_values:
model_filter = {'random_no': target_value[0], 'random_yes': target_value[1]}
possible_models = get_models_by_filter(model_filter, model_infos)
target_model = None
for pseudo_model_list in pseudo_models[pseudo_model_name]:
for possible_model in possible_models:
if possible_model in pseudo_model_list:
#print(possible_model, pseudo_model_list)
target_model = possible_model
break
#print(target_model)
if target_model is not None:
val = 0
for target_dataset in test_collections[test_collection_name]:
evaluation = (target_model, target_dataset.name)
val += getattr(model_evaluation.predictions[evaluation].evaluation, target_score)
y_vals.append(val/len(test_collections[test_collection_name]))
nonzero_val = target_value[0]
if nonzero_val == 0 or nonzero_val == 1:
nonzero_val = target_value[1]
x_vals.append(nonzero_val)
 
ax.plot(x_vals, y_vals, label=pseudo_model_name)
 
val = 0
for target_dataset in test_collections[test_collection_name]:
val += getattr(model_evaluation.predictions[('HandWashingDeepConvLSTMA_trunc_01.pt', target_dataset.name)].evaluation, target_score)
ax.axhline(val/len(test_collections[test_collection_name]), color='red', linestyle='--', label='genereal model')
 
val = 0
for target_dataset in test_collections[test_collection_name]:
val += getattr(model_evaluation.predictions[('synthetic_inc_01.pt', target_dataset.name)].evaluation, target_score)
ax.axhline(val/len(test_collections[test_collection_name]), color='green', linestyle='--', label='gt inc model')
 
 
 
# plt.xticks(rotation=45, ha='right')
try:
fig.legend(loc='lower right')
except NameError:
pass
 
return ax
 
 
def plot_randomized_pseudo_label_data(target_dataset, target_score, target_values, model_infos, collection_name):
fig, ax = plt.subplots(figsize=(9,5))
 
ax.set_title(target_score)
ax.set_ylabel(target_score)
ax.set_xlabel('evaluation reliability')
 
 
 
collection = training_runs[collection_name]
pseudo_models = dict()
for pseudo_model_name in relevant_models:
if pseudo_model_name not in pseudo_models:
pseudo_models[pseudo_model_name] = []
pseudo_models[pseudo_model_name].append(get_models_with_generators(pseudo_model_settings[pseudo_model_name], training_manager.get_all_information(), collection))
 
for pseudo_model_name in relevant_models:
y_vals = []
x_vals = []
for target_value in target_values:
model_filter = {'random_no': target_value[0], 'random_yes': target_value[1]}
possible_models = get_models_by_filter(model_filter, model_infos)
target_model = None
for pseudo_model_list in pseudo_models[pseudo_model_name]:
for possible_model in possible_models:
if possible_model in pseudo_model_list:
#print(possible_model, pseudo_model_list)
target_model = possible_model
break
#print(target_model)
if target_model is not None:
# evaluation = (target_model, target_dataset)
# print(training_manager.get_all_information()[target_model])
val = training_manager.get_all_information()[target_model]['pseudo_evaluation'][target_score]
 
#val = getattr(model_evaluation.predictions[evaluation].evaluation, target_score)
y_vals.append(val)
if np.isnan(val):
print(target_model)
nonzero_val = target_value[0]
if nonzero_val == 0 or nonzero_val == 1:
nonzero_val = target_value[1]
x_vals.append(nonzero_val)
 
ax.plot(x_vals, y_vals, label=pseudo_model_name)
 
val = getattr(model_evaluation.predictions[('HandWashingDeepConvLSTMA_trunc_01.pt', target_dataset)].evaluation, target_score)
ax.axhline(val, color='red', linestyle='--', label='genereal model')
val = getattr(model_evaluation.predictions[('synthetic_inc_01.pt', target_dataset)].evaluation, target_score)
ax.axhline(val, color='green', linestyle='--', label='gt inc model')
 
 
 
# plt.xticks(rotation=45, ha='right')
try:
fig.legend(loc='lower right')
except NameError:
pass
 
return ax
 
 
def plot_all_randomized_averages(collection_name, target_values, model_info, run_name):
plot_randomized_average(collection_name, 'specificity', target_values, model_info, run_name)
plot_randomized_average(collection_name, 'sensitivity', target_values, model_info, run_name)
plot_randomized_average(collection_name, 'f1', target_values, model_info, run_name)
plot_randomized_average(collection_name, 's1', target_values, model_info, run_name)
 
def plot_all_randomized_values(collection_name, target_values, model_info, run_name):
plot_randomized_pseudo_label_data(collection_name, 'specificity', target_values, model_info, run_name)
plot_randomized_pseudo_label_data(collection_name, 'sensitivity', target_values, model_info, run_name)
plot_randomized_pseudo_label_data(collection_name, 'f1', target_values, model_info, run_name)
plot_randomized_pseudo_label_data(collection_name, 's1', target_values, model_info, run_name)
 
 
```
 
%% Cell type:code id:774c4464 tags:
 
``` python
model_info = training_manager.get_all_information()
relevant_model_info = dict()
for model in training_runs['pseudo_randomeval_01_run1']:
relevant_model_info[model] = model_info[model]
```
 
%% Cell type:code id:201c951e tags:
 
``` python
#zero_no_values = [(info['random_no'], info['random_yes']) for info in relevant_model_info.values() if info['random_no'] == 0]
#zero_yes_values = [(info['random_no'], info['random_yes']) for info in relevant_model_info.values() if info['random_yes'] == 0]
 
#zero_no_values = sorted(set(zero_no_values), key=lambda pair: pair[1])
#zero_yes_values = sorted(set(zero_yes_values), key=lambda pair: pair[0])
 
one_no_values = [(info['random_no'], info['random_yes']) for info in relevant_model_info.values() if info['random_no'] == 1]
one_yes_values = [(info['random_no'], info['random_yes']) for info in relevant_model_info.values() if info['random_yes'] == 1]
 
one_no_values = sorted(set(one_no_values), key=lambda pair: pair[1])
one_yes_values = sorted(set(one_yes_values), key=lambda pair: pair[0])
 
```
 
%% Cell type:code id:e2c817ad tags:
 
``` python
print(one_no_values)
print(one_yes_values)
```
 
%% Cell type:code id:b9c119d9 tags:
 
``` python
#relevant_models = list(pseudo_model_settings.keys())
relevant_models = ['synthetic', 'allnoise_correctedhwgt', 'scope_corrected', 'allnoise_corrected', 'allnoise_correctedscore', 'allnoise_correctbydeepconvfilter', 'allnoise_correctbyfcndaefilter', 'allnoise_correctbyconvlstm3filter', 'alldeepconv_correctbyconvlstm3filter', 'alldeepconv_correctbyconvlstm3filter2', 'alldeepconv_correctbyconvlstm3filter3', 'alldeepconv_correctbyconvlstm3filter4', 'alldeepconv_correctbyconvlstm3filter5', 'alldeepconv_correctbyconvlstm3filter6']
relevant_models = ['synthetic', 'allnoise_correctedhwgt', 'scope_corrected', 'allnoise_corrected', 'allnoise_correctedscore', 'allnoise_correctbydeepconvfilter', 'allnoise_correctbyfcndaefilter', 'allnoise_correctbyconvlstm3filter', 'alldeepconv_correctbyconvlstm3filter', 'alldeepconv_correctbyconvlstm3filter6']
```
 
%% Cell type:code id:c135f442 tags:
 
``` python
# plot_randomized('01_generated_1', 'f1', one_no_values, relevant_model_info, 'pseudo_randomeval_01_run1')
# plot_randomized('01_generated_1', 'f1', one_yes_values, relevant_model_info, 'pseudo_randomeval_01_run1')
plot_all_randomized_averages('01_test', one_no_values, relevant_model_info, 'pseudo_randomeval_01_run1')
plot_all_randomized_averages('01_test', one_yes_values, relevant_model_info, 'pseudo_randomeval_01_run1')
#plt.subplots_adjust(hspace=0.4)
plt.subplots_adjust(right=0.85)
```
 
%% Cell type:code id:b5acd52b tags:
 
``` python
plot_all_randomized_values('01_generated_0', one_no_values, relevant_model_info,'pseudo_randomeval_01_run1')
plot_all_randomized_values('01_generated_0', one_yes_values, relevant_model_info,'pseudo_randomeval_01_run1')
#plt.subplots_adjust(hspace=0.4)
plt.subplots_adjust(right=0.85)
```
 
%% Cell type:code id:42ac6954 tags:
 
``` python
def do_random_plots(dataset_name, target_value, values, model_info):
fig, ax = plt.subplots()
plot_randomized(dataset_name, target_value, values, label_filters['allnoise correctedscore'], model_info, label='allnoise correctedscore', ax=ax)
plot_randomized(dataset_name, target_value, values, label_filters['allnoise correctedscore_flatten'], model_info, label='allnoise correctedscore_flatten', ax=ax, plot_baseline=False)
# plt.legend(loc = 'lower left', bbox_to_anchor=(0.5, 0))
plt.legend(loc="best", ncol=2)
#fig.legend()
 
do_random_plots('01_generated_0', 'specificity', one_yes_values, relevant_model_info)
do_random_plots('01_generated_0', 'sensitivity', one_yes_values, relevant_model_info)
do_random_plots('01_generated_0', 'f1', one_yes_values, relevant_model_info)
do_random_plots('01_generated_0', 's1', one_yes_values, relevant_model_info)
```
 
%% Cell type:code id:b63d014c tags:
 
``` python
for key in model_evaluation.predictions.keys():
if 'random' in key:
print(key)
```
 
%% Cell type:code id:75573056 tags:
 
``` python
fig, ax = plt.subplots()
plot_randomized('01_generated_0', 'specificity', one_no_values, {'corrected': False, 'scope': False, 'exneut': False}, relevant_model_info_2, label='all', ax=ax)
plot_randomized('01_generated_0', 'specificity', one_no_values, {'corrected': False, 'scope': True, 'exneut': False}, relevant_model_info_2, label='scope', ax=ax, plot_baseline=False)
plot_randomized('01_generated_0', 'specificity', one_no_values, {'corrected': False, 'scope': True, 'exneut': True}, relevant_model_info_2, label='no neut', ax=ax, plot_baseline=False)
plot_randomized('01_generated_0', 'specificity', one_no_values, {'corrected': True, 'scope': False, 'exneut': False}, relevant_model_info_2, label='corrected', ax=ax, plot_baseline=False)
plot_randomized('01_generated_0', 'specificity', one_no_values, {'corrected': True, 'scope': True, 'exneut': True}, relevant_model_info_2, label='corr+scope', ax=ax, plot_baseline=False)
plot_randomized('01_generated_0', 'specificity', one_no_values, {'corrected': True, 'scope': True, 'exneut': True, 'correctedhwgt': True}, relevant_model_info_2, label='corr hw gt', ax=ax, plot_baseline=False)
plt.legend(loc="best", ncol=2)
 
 
fig, ax = plt.subplots()
plot_randomized('01_generated_0', 'sensitivity', one_no_values, {'corrected': False, 'scope': False, 'exneut': False}, relevant_model_info_2, label='all', ax=ax)
plot_randomized('01_generated_0', 'sensitivity', one_no_values, {'corrected': False, 'scope': True, 'exneut': False}, relevant_model_info_2, label='scope', ax=ax, plot_baseline=False)
plot_randomized('01_generated_0', 'sensitivity', one_no_values, {'corrected': False, 'scope': True, 'exneut': True}, relevant_model_info_2, label='no neut', ax=ax, plot_baseline=False)
plot_randomized('01_generated_0', 'sensitivity', one_no_values, {'corrected': True, 'scope': False, 'exneut': False}, relevant_model_info_2, label='corrected', ax=ax, plot_baseline=False)
plot_randomized('01_generated_0', 'sensitivity', one_no_values, {'corrected': True, 'scope': True, 'exneut': True}, relevant_model_info_2, label='corr+scope', ax=ax, plot_baseline=False)
plot_randomized('01_generated_0', 'sensitivity', one_no_values, {'corrected': True, 'scope': True, 'exneut': True, 'correctedhwgt': True}, relevant_model_info_2, label='corr hw gt', ax=ax, plot_baseline=False)
plt.legend(loc="best", ncol=2)
 
 
fig, ax = plt.subplots()
plot_randomized('01_generated_0', 'f1', one_no_values, {'corrected': False, 'scope': False, 'exneut': False}, relevant_model_info_2, label='all', ax=ax)
plot_randomized('01_generated_0', 'f1', one_no_values, {'corrected': False, 'scope': True, 'exneut': False}, relevant_model_info_2, label='scope', ax=ax, plot_baseline=False)
plot_randomized('01_generated_0', 'f1', one_no_values, {'corrected': False, 'scope': True, 'exneut': True}, relevant_model_info_2, label='no neut', ax=ax, plot_baseline=False)
plot_randomized('01_generated_0', 'f1', one_no_values, {'corrected': True, 'scope': False, 'exneut': False}, relevant_model_info_2, label='corrected', ax=ax, plot_baseline=False)
plot_randomized('01_generated_0', 'f1', one_no_values, {'corrected': True, 'scope': True, 'exneut': True}, relevant_model_info_2, label='corr+scope', ax=ax, plot_baseline=False)
plot_randomized('01_generated_0', 'f1', one_no_values, {'corrected': True, 'scope': True, 'exneut': True, 'correctedhwgt': True}, relevant_model_info_2, label='corr hw gt', ax=ax, plot_baseline=False)
plt.legend(loc="best", ncol=2)
 
 
fig, ax = plt.subplots()
plot_randomized('01_generated_0', 's1', one_no_values, {'corrected': False, 'scope': False, 'exneut': False}, relevant_model_info_2, label='all', ax=ax)
plot_randomized('01_generated_0', 's1', one_no_values, {'corrected': False, 'scope': True, 'exneut': False}, relevant_model_info_2, label='scope', ax=ax, plot_baseline=False)
plot_randomized('01_generated_0', 's1', one_no_values, {'corrected': False, 'scope': True, 'exneut': True}, relevant_model_info_2, label='no neut', ax=ax, plot_baseline=False)
plot_randomized('01_generated_0', 's1', one_no_values, {'corrected': True, 'scope': False, 'exneut': False}, relevant_model_info_2, label='corrected', ax=ax, plot_baseline=False)
plot_randomized('01_generated_0', 's1', one_no_values, {'corrected': True, 'scope': True, 'exneut': True}, relevant_model_info_2, label='corr+scope', ax=ax, plot_baseline=False)
plot_randomized('01_generated_0', 's1', one_no_values, {'corrected': True, 'scope': True, 'exneut': True, 'correctedhwgt': True}, relevant_model_info_2, label='corr hw gt', ax=ax, plot_baseline=False)
plt.legend(loc="best", ncol=2)
```
 
%% Cell type:code id:a0d208ea tags:
 
``` python
fig, ax = plt.subplots()
plot_randomized('01_generated_1', 'specificity', one_yes_values, {'corrected': False, 'scope': False, 'exneut': False}, relevant_model_info_2, label='all', ax=ax)
plot_randomized('01_generated_1', 'specificity', one_yes_values, {'corrected': False, 'scope': True, 'exneut': False}, relevant_model_info_2, label='scope', ax=ax, plot_baseline=False)
plot_randomized('01_generated_1', 'specificity', one_yes_values, {'corrected': False, 'scope': True, 'exneut': True}, relevant_model_info_2, label='no neut', ax=ax, plot_baseline=False)
plot_randomized('01_generated_1', 'specificity', one_yes_values, {'corrected': True, 'scope': False, 'exneut': False}, relevant_model_info_2, label='corrected', ax=ax, plot_baseline=False)
plot_randomized('01_generated_1', 'specificity', one_yes_values, {'corrected': True, 'scope': True, 'exneut': True}, relevant_model_info_2, label='corr+scope', ax=ax, plot_baseline=False)
plt.legend(loc="best", ncol=2)
 
fig, ax = plt.subplots()
plot_randomized('01_generated_1', 'sensitivity', one_yes_values, {'corrected': False, 'scope': False, 'exneut': False}, relevant_model_info_2, label='all', ax=ax)
plot_randomized('01_generated_1', 'sensitivity', one_yes_values, {'corrected': False, 'scope': True, 'exneut': False}, relevant_model_info_2, label='scope', ax=ax, plot_baseline=False)
plot_randomized('01_generated_1', 'sensitivity', one_yes_values, {'corrected': False, 'scope': True, 'exneut': True}, relevant_model_info_2, label='no neut', ax=ax, plot_baseline=False)
plot_randomized('01_generated_1', 'sensitivity', one_yes_values, {'corrected': True, 'scope': False, 'exneut': False}, relevant_model_info_2, label='corrected', ax=ax, plot_baseline=False)
plot_randomized('01_generated_1', 'sensitivity', one_yes_values, {'corrected': True, 'scope': True, 'exneut': True}, relevant_model_info_2, label='corr+scope', ax=ax, plot_baseline=False)
plt.legend(loc="best", ncol=2)
 
fig, ax = plt.subplots()
plot_randomized('01_generated_1', 'f1', one_yes_values, {'corrected': False, 'scope': False, 'exneut': False}, relevant_model_info_2, label='all', ax=ax)
plot_randomized('01_generated_1', 'f1', one_yes_values, {'corrected': False, 'scope': True, 'exneut': False}, relevant_model_info_2, label='scope', ax=ax, plot_baseline=False)
plot_randomized('01_generated_1', 'f1', one_yes_values, {'corrected': False, 'scope': True, 'exneut': True}, relevant_model_info_2, label='no neut', ax=ax, plot_baseline=False)
plot_randomized('01_generated_1', 'f1', one_yes_values, {'corrected': True, 'scope': False, 'exneut': False}, relevant_model_info_2, label='corrected', ax=ax, plot_baseline=False)
plot_randomized('01_generated_1', 'f1', one_yes_values, {'corrected': True, 'scope': True, 'exneut': True}, relevant_model_info_2, label='corr+scope', ax=ax, plot_baseline=False)
plt.legend(loc="best", ncol=2)
 
fig, ax = plt.subplots()
plot_randomized('01_generated_1', 's1', one_yes_values, {'corrected': False, 'scope': False, 'exneut': False}, relevant_model_info_2, label='all', ax=ax)
plot_randomized('01_generated_1', 's1', one_yes_values, {'corrected': False, 'scope': True, 'exneut': False}, relevant_model_info_2, label='scope', ax=ax, plot_baseline=False)
plot_randomized('01_generated_1', 's1', one_yes_values, {'corrected': False, 'scope': True, 'exneut': True}, relevant_model_info_2, label='no neut', ax=ax, plot_baseline=False)
plot_randomized('01_generated_1', 's1', one_yes_values, {'corrected': True, 'scope': False, 'exneut': False}, relevant_model_info_2, label='corrected', ax=ax, plot_baseline=False)
plot_randomized('01_generated_1', 's1', one_yes_values, {'corrected': True, 'scope': True, 'exneut': True}, relevant_model_info_2, label='corr+scope', ax=ax, plot_baseline=False)
plt.legend(loc="best", ncol=2)
```
 
%% Cell type:code id:1a23e507 tags:
 
``` python
relevant_model_info_2
```
 
%% Cell type:code id:a6ec2f44 tags:
 
``` python
print(get_models_by_filter({'corrected': True, 'scope': True, 'exneut': True, 'correctedhwgt': True}, relevant_model_info_2))
```
 
%% Cell type:code id:d2fe193f tags:
 
``` python
relevant_model_info_2['pseudo_randomeval_01_run4_pseudo_collection_noneut_correctedhwgt_rn0ry0.pt']
```
 
%% Cell type:code id:af2adf1e tags:
 
``` python
np.around(0.7000000000000002, decimals=3)
```
 
%% Cell type:code id:35d4d30a tags:
 
``` python
collection = training_runs['pseudo_randomeval_01_run1']
get_models_with_generators(pseudo_model_settings['allnoise_correctedhwgt'], training_manager.get_all_information(), collection)[-1]
training_manager.get_all_information()['pseudo_randomeval_01_run1_allnoise_correctedhwgt_rn0ry2.pt']['pseudo_evaluation']['f1']
```
 
%% Cell type:code id:a6e374de tags:
 
``` python
training_manager.get_all_information()
```
 
%% Cell type:code id:5487fb8e tags:
 
``` python
pseudo_model_settings.keys()
```
 
%% Cell type:code id:f1f82e42 tags:
 
``` python
plot_randomized_pseudo_label_data('01_generated_0', 'f1', one_no_values, relevant_model_info, 'pseudo_randomeval_01_run1')
```
 
%% Cell type:markdown id:4a4c1773 tags:
 
# Tables for thesis
 
%% Cell type:code id:ab2b1b29 tags:
 
``` python
https://stackoverflow.com/questions/32370402/giving-a-column-multiple-indexes-headers
```
 
%% Cell type:code id:8c93dfa8 tags:
 
``` python
```
 
%% Cell type:code id:9c65338a tags:
 
``` python
```
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment