add code for processing data and model cards

app.py

@@ -1,15 +1,25 @@
 import gradio as gr
 import yaml
-from compliance_analysis import run_compliance_analysis_on_project
+from pathlib import Path
+from compliance_analysis import run_compliance_analysis_on_project, run_compliance_analysis_on_data, run_compliance_analysis_on_model
 
 def process_files(files):
     results = []
     for file in files:
         with open(file.name, 'r') as f:
             content = f.read()
-        project_cc_yaml = yaml.safe_load(content)
-        msg = run_compliance_analysis_on_project(project_cc_yaml)
-        results.append(msg)            
+        if Path(file.name).name == "project_cc.yaml":
+            project_cc_yaml = yaml.safe_load(content)
+            msg = run_compliance_analysis_on_project(project_cc_yaml)
+            results.append(msg)            
+        # if Path(file.name).name == "data_cc.yaml":
+        #     data_cc_yaml = yaml.safe_load(content)
+        #     msg = run_compliance_analysis_on_data(data_cc_yaml)
+        #     results.append(msg)        
+        # if Path(file.name).name == "model_cc.yaml":
+        #     model_cc_yaml = yaml.safe_load(content)
+        #     msg = run_compliance_analysis_on_model(model_cc_yaml)
+        #     results.append(msg)
             
     return results
 

compliance_analysis.py

@@ -76,45 +76,39 @@ def run_compliance_analysis_on_project(project_cc_yaml):
     return msg
 
 def run_compliance_analysis_on_data(data_cc_yaml):
-    # Do this by examining any and all Data CCs too
-    data_cc_yaml = yaml.safe_load(data_cc_yaml)
-    filename = data_cc_yaml
     
     for key, value in data_cc_yaml['data_and_data_governance']:
         if not value:
-            msg = (f"Because of the dataset represented by {filename}, this high-risk AI system fails the data and data governance requirements under Article 10.")
+            msg = (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:
-            msg = (f"Because of the dataset represented by {filename}, this high-risk AI system fails the technical documentation requirements under Article 11.")
+            msg = (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:
-            msg = (f"Because of the dataset represented by {filename}, this high-risk  AI system fails the transparency requirements under Article 13.")
+            msg = (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:
-            msg = (f"Because of the dataset represented by {filename}, this high-risk  AI system fails the quality management requirements under Article 17.")
+            msg = (f"Because of the dataset represented by , this high-risk  AI system fails the quality management requirements under Article 17.")
 
     return msg
     
-def run_compliance_analysis_on_data(model_cc_yaml):
-    
-    model_cc_yaml = yaml.safe_load(model_cc_yaml)
-    filename - model_cc_yaml
+def run_compliance_analysis_on_model(model_cc_yaml):
     
     for key, value in model_cc_yaml['risk_management_system']:
         if not value:
-            msg = (f"Because of the model represented by {filename}, this high-risk AI system fails the risk management requirements under Article 9.")
+            msg = (f"Because of the model represented by , this high-risk AI system fails the risk management requirements under Article 9.")
     for key, value in data_cc_yaml['technical_documentation']:
         if not value:
-            msg = (f"Because of the model represented by {filename}, this high-risk AI system fails the technical documentation requirements under Article 11.")
+            msg = (f"Because of the model 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:
-            msg = (f"Because of the model represented by {filename}, this high-risk  AI system fails the transparency requirements under Article 13.")
+            msg = (f"Because of the model represented by , this high-risk  AI system fails the transparency requirements under Article 13.")
     for key, value in data_cc_yaml['accuracy_robustness_cybersecurity']:
         if not value:
-            msg = (f"Because of the model represented by {filename}, this high-risk  AI system fails the quality management requirements under Article 15.")
+            msg = (f"Because of the model represented by , this high-risk  AI system fails the quality management requirements under Article 15.")
     for key, value in data_cc_yaml['quality_management_system']:
         if not value:
-            msg = (f"Because of the model represented by {filename}, this high-risk  AI system fails the quality management requirements under Article 17.")
+            msg = (f"Because of the model represented by , this high-risk  AI system fails the quality management requirements under Article 17.")
    
     return msg
 

data_cc.yaml

@@ -68,13 +68,13 @@ data_and_data_governance:
   personal_data_deletion: # Art. 10(5)(e)
     verbose: 'Where special categories of personal data have been used to ensure the detection and correction 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, the special categories of personal data were deleted once the bias was corrected or the personal data reached the end of its retention period (whichever came first)'
     value: !!bool false
-  personal_data_necessary: # Art. 10(5)(f)
+  personal_data_necessary_105f: # Art. 10(5)(f)
     verbose: 'Where special categories of personal data have been used to ensure the detection and correction 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, the records of processing activities pursuant to Regulations (EU) 2016/679 and (EU) 2018/1725 and Directive (EU) 2016/680 include the reasons why the processing of special categories of personal data was strictly necessary to detect and correct biases, and why that objective could not be achieved by processing other data'
     value: !!bool false
 
 technical_documentation:
   general_description: # Art. 11; Annex IV(2)(d)
-    verbose: 'Dataset carries technical documention, such as a dataseet, including a general description of the dataset." 
+    verbose: 'Dataset carries technical documention, such as a dataseet, including a general description of the dataset.'
     value: !!bool false
   provenance: # Art. 11; Annex IV(2)(d)
     verbose: 'Dataset carries technical documention, such as a dataseet, including information about its provenance' 

model_cc.yaml

@@ -13,7 +13,7 @@ risk_management_system:
     value: !!bool false
   testing_compliance: # Art. 9(6)
     verbose: 'Testing to ensure model complies with Act'
-   value: !!bool false
+    value: !!bool false
   testing_benchmark: # Art. 9(8)
     verbose: 'Testing against prior defined metrics appropriate to intended purpose'
     value: !!bool false
@@ -87,7 +87,7 @@ accuracy_robustness_cybersecurity:
     verbose: 'Model is designed and developed to achieve appropriate level of accuracy'
     value: !!bool false
   robustiness: # Art. 15(1)
-    verbose 'Model is designed and developed to achieve appropriate level of robustness'
+    verbose: 'Model is designed and developed to achieve appropriate level of robustness'
     value: !!bool false
   cybersecurity:  # Art. 15(1)
     verbose: 'Model is designed and developed to achieve appropriate level of cybersecurity'