update run_compliance_analysis_on_data

.gitignore

@@ -1,5 +1,4 @@
-__pycache__
+__pycache__/
 compliancecards.code-workspace
 .ipynb_checkpoints
 *.ipynb
-__pycache__/*

compliance_analysis.py

@@ -46,10 +46,10 @@ def check_overall_compliance(cards):
     # for each model_cc and data_cc - run analysis with ref to project_cc
     dispositive_variables = run_compliance_analysis_on_project(dispositive_variables, project_cc)
 
-    # for card in cards['data_files']:
-    #     with open(card, 'r') as data_filepath:
-    #         data_cc = yaml.safe_load(data_filepath.read())
-    #         dispositive_variables = run_compliance_analysis_on_data(dispositive_variables, data_cc_yaml)
+    for card in cards['data_files']:
+        with open(card, 'r') as data_filepath:
+            data_cc = yaml.safe_load(data_filepath.read())
+            dispositive_variables = run_compliance_analysis_on_data(dispositive_variables, data_cc)
             
     # for card in cards['model_files']:
     #     with open(card, 'r') as model_filepath:
@@ -156,20 +156,10 @@ def run_compliance_analysis_on_project(dispositive_variables, project_cc_yaml):
 
 def run_compliance_analysis_on_data(dispositive_variables, data_cc_yaml): 
     
-    # # TODO: we probably have to pass ai_project_type and project_intended_purpose into this function
-    # if dispositive_variables['ai_project_type']["high_risk_ai_system"] == True:
-    #     for value in data_cc_yaml['high_risk_ai_systems']['data_and_data_governance']:
-    #         if not value:
-    #             dispositive_variables['msg'].append(f"Because of the dataset represented by , this high-risk AI system fails the data and data governance requirements under Article 10.")
-    #     for key, value in data_cc_yaml['technical_documentation']:
-    #         if not value:
-    #             dispositive_variables['msg'].append(f"Because of the dataset represented by , this high-risk AI system fails the technical documentation requirements under Article 11.")
-    #     for key, value in data_cc_yaml['transparency_and_provision_of_information_to_deployers']:
-    #         if not value:
-    #             dispositive_variables['msg'].append(f"Because of the dataset represented by , this high-risk  AI system fails the transparency requirements under Article 13.")
-    #     for key, value in data_cc_yaml['quality_management_system']:
-    #         if not value:
-    #             dispositive_variables['msg'].append(f"Because of the dataset represented by , this high-risk  AI system fails the quality management requirements under Article 17.")
+    if dispositive_variables['ai_project_type']["high_risk_ai_system"] == True:
+        for key in data_cc_yaml['high_risk_ai_systems']:
+            if data_cc_yaml['high_risk_ai_systems'][f'{key}']['value'] == True:
+                dispositive_variables['msg'].append(f"This high-risk AI system fails the {key} requirements under {data_cc_yaml['high_risk_ai_systems'][f'{key}']['article']}.")
 
     if dispositive_variables['ai_project_type']["gpai_model"] == True:
         for value in data_cc_yaml['gpai_requirements']:

data_cc.yaml

@@ -83,14 +83,14 @@ high_risk_ai_systems:
     article: 'Art. 10(2)(e)'
     verbose: 'The dataset has been subject to data governance and management practices that include an assessment of the availability, quantity and suitability of the data sets that are needed'
     value: !!bool false
-  ddata_and_data_governance_ata_bias_examination: 
+  data_and_data_governance_ata_bias_examination: 
     article: 'Art. 10(2)(f)'
     verbose: 'The dataset has been subject to data governance and management practices that include an examination of possible biases that are likely to affect the health and safety of persons, have a negative impact on fundamental rights or lead to discrimination prohibited under Union law, especially where data outputs influence inputs for future operations' 
     value: !!bool false
   data_and_data_governance_data_and_data_governance_data_bias_mitigation: 
     article: 'Art. 10(2)(g)'
     verbose: 'The dataset has been subject to data governance and management practices that include appropriate measures to detect, prevent and mitigate possible biases that are likely to affect the health and safety of persons, have a negative impact on fundamental rights or lead to discrimination prohibited under Union law, especially where data outputs influence inputs for future operations'
-    value: !!bool false
+    value: !!bool true
   data_and_data_governance_data_compliance: 
     article: 'Art. 10(2)(h)'
     verbose: 'The dataset has been subject to data governance and management practices that include identification of relevant data gaps or shortcomings that prevent compliance with this Regulation, and how those gaps and shortcomings can be addressed'