def test_write_output(self):
print('++ テスト開始')
aaa = AITManifestGenerator('./')
aaa.set_ait_name("set_ait_name")
aaa.set_ait_description("set_ait_description")
aaa.set_ait_author("set_ait_author")
aaa.set_ait_email("set_ait_email")
aaa.set_ait_version("0.1")
aaa.set_ait_quality("set_ait_quality")
aaa.set_ait_reference("set_ait_reference")
aaa.add_ait_inventories('name1', 'type1', 'description1', ['csv'],
'schema1')
aaa.add_ait_inventories('name2', 'type2', 'description2',
['gz', 'zip'], 'schema')
aaa.add_ait_parameters('name1', 'type1', 'description1',
'default_val1')
aaa.add_ait_parameters('name2', 'type2', 'description2')
aaa.add_ait_measures('name1', 'type1', 'description1', 'structure1')
aaa.add_ait_measures('name2', 'type2', 'description2', 'structure2')
aaa.add_ait_resources('name1', 'type1', 'description1')
aaa.add_ait_resources('name2', 'type2', 'description2')
aaa.add_ait_downloads('name1', 'description1')
aaa.add_ait_downloads('name2', 'description2')
aaa.write()
print('++ テスト終了')
name='mnist_type',
type_='str',
description=
'train = Training_data, test = test_data, validation = validation_data',
default_val='train')
manifest_genenerator.add_ait_measures(
name='average',
type_='float',
description='Average number of NG output',
structure='single',
min='0',
max='1')
manifest_genenerator.add_ait_resources(name='result',
type_='table',
description='number of NG output')
manifest_genenerator.add_ait_downloads(name='Log', description='AIT_log')
manifest_genenerator.add_ait_downloads(name='DeepLog',
description='deep_saucer_log')
manifest_path = manifest_genenerator.write()
# In[9]:
#########################################
# area:create input
# should edit
#########################################
if not is_ait_launch:
from ait_sdk.common.files.ait_input_generator import AITInputGenerator
input_generator = AITInputGenerator(manifest_path)
input_generator.add_ait_inventories(
name='mnist_dataset', value='mnist_dataset/mnist_dataset.zip')
type_='float',
description='privileged classes',
default_val='25.0')
manifest_genenerator.add_ait_measures(
name='mean_difference',
type_='float',
description='mean difference of metric fairness',
structure='single',
min='-1',
max='1')
manifest_genenerator.add_ait_resources(
name='metric_fairness_plot',
type_='picture',
description=
'base rates polt of privileged_groups and unprivileged_groups')
manifest_genenerator.add_ait_downloads(name='Log', description='AITLog')
manifest_path = manifest_genenerator.write()
# In[9]:
#########################################
# area:create input
# should edit
#########################################
if not is_ait_launch:
from ait_sdk.common.files.ait_input_generator import AITInputGenerator
input_generator = AITInputGenerator(manifest_path)
input_generator.add_ait_inventories(name='Data',
value='german_credit/german.csv')
input_generator.set_ait_params(name='protected_attribute', value='age')
input_generator.set_ait_params(name='privileged_classes', value='25.0')
structure='single',
min='0',
max='1')
manifest_genenerator.add_ait_resources(
name='coefficient_of_determination_matrix',
type_='table',
description='決定係数の結果をまとめた表')
manifest_genenerator.add_ait_resources(
name='correlation_coefficient_1to1',
type_='picture',
description='説明変数と目的変数の相関グラフ\nファイル名は{目的変数}-{説明変数}.png')
manifest_genenerator.add_ait_resources(
name='correlation_coefficient_1to2',
type_='picture',
description='説明変数2つと目的変数の3次元相関グラフ\nファイル名は{目的変数}-{説明変数1}-{説明変数2}.png')
manifest_genenerator.add_ait_downloads(name='Log', description='AIT実行ログ')
manifest_genenerator.add_ait_downloads(name='predictive_value',
description='予測値')
manifest_path = manifest_genenerator.write()
# In[ ]:
#########################################
# area:create input
# should edit
#########################################
if not is_ait_launch:
from ait_sdk.common.files.ait_input_generator import AITInputGenerator
input_generator = AITInputGenerator(manifest_path)
input_generator.add_ait_inventories('trained_model',
'trained_model/model_1.h5')
min='0',
max='1')
manifest_genenerator.add_ait_measures(
name='train_accuracy',
type_='float',
description='accuracy predicted of train',
structure='single',
min='0',
max='1')
manifest_genenerator.add_ait_resources(name='all_label_accuracy_csv',
type_='table',
description='accuracy of all label')
manifest_genenerator.add_ait_resources(name='all_label_accuracy_png',
type_='picture',
description='accuracy of all label')
manifest_genenerator.add_ait_downloads(name='Log', description='AIT_log')
manifest_genenerator.add_ait_downloads(
name='each_label_accuracy', description='accuracy of each label')
manifest_genenerator.add_ait_downloads(
name='each_picture', description='predict of each picture')
manifest_path = manifest_genenerator.write()
# In[9]:
#########################################
# area:create input
# should edit
#########################################
if not is_ait_launch:
from ait_sdk.common.files.ait_input_generator import AITInputGenerator
input_generator = AITInputGenerator(manifest_path)
structure='sequence',
min='0',
max='1')
manifest_genenerator.add_ait_measures(
name='coverage_rate_combination',
type_='float',
description='coverage of select combination.',
structure='single',
min='0',
max='1')
manifest_genenerator.add_ait_resources(
name='test_case_generator',
type_='table',
description='generate coverage increase data.')
manifest_genenerator.add_ait_downloads(
name='heatmap',
description='the heat map of the coverage as an HTML file.')
manifest_genenerator.add_ait_downloads(
name='abs_dataset',
description='the created input data by the manipulation as h5 file.')
manifest_genenerator.add_ait_downloads(name='Log', description='AITLog')
manifest_path = manifest_genenerator.write()
# In[9]:
#########################################
# area:create input
# should edit
#########################################
if not is_ait_launch:
from ait_sdk.common.files.ait_input_generator import AITInputGenerator
description='F−measure for each class.',
structure='sequence',
min='0',
max='1')
manifest_genenerator.add_ait_resources(name='ConfusionMatrixHeatmap',
type_='picture',
description='混同行列(ヒートマップ)')
manifest_genenerator.add_ait_resources(name='ROC-curve',
type_='picture',
description='ROC曲線')
manifest_genenerator.add_ait_resources(name='NGPredictImages',
type_='picture',
description='推論NGとなった画像の一覧を、正解ラベルの枚数分だけ出力する')
manifest_genenerator.add_ait_downloads(name='Log',
description='AIT実行ログ')
manifest_genenerator.add_ait_downloads(name='ConfusionMatrixCSV',
description='混同行列')
manifest_genenerator.add_ait_downloads(name='PredictionResult',
description='ID,正解ラベル,推論結果確率(ラベル毎)')
manifest_path = manifest_genenerator.write()
# In[9]:
#########################################
# area:create input
# should edit
#########################################
'Classification of different attributes related to autonomous driving scenarios',
['csv'], 'https://bdd-data.berkeley.edu/')
manifest_genenerator.add_ait_parameters(
'attribute_no', 'int',
'Total number of attribute for distibution analysis', '6')
manifest_genenerator.add_ait_parameters(
'dimension', 'int',
'Dimensions of how many attributes to combine for distibution analysis',
'2')
manifest_genenerator.add_ait_resources(
'distibution_csv', 'table',
'Table of distribution for each combination')
manifest_genenerator.add_ait_resources(
'distibution_plot', 'picture',
'Plot of distribution for each combination')
manifest_genenerator.add_ait_downloads('Log', 'AITLog')
manifest_path = manifest_genenerator.write()
# In[6]:
#########################################
# area:create input
# should edit
#########################################
if not is_ait_launch:
from ait_sdk.common.files.ait_input_generator import AITInputGenerator
input_generator = AITInputGenerator(manifest_path)
input_generator.add_ait_inventories('Data', 'BDD_data/BDD_labels_2036.csv')
input_generator.set_ait_params('attribute_no', '7')
input_generator.set_ait_params('dimension', '2')
input_generator.write()