For Ezi

.idea/.gitignore

@@ -0,0 +1,8 @@
+# Default ignored files
+/shelf/
+/workspace.xml
+# Editor-based HTTP Client requests
+/httpRequests/
+# Datasource local storage ignored files
+/dataSources/
+/dataSources.local.xml

.idea/inspectionProfiles/Project_Default.xml

@@ -0,0 +1,30 @@
+<component name="InspectionProjectProfileManager">
+  <profile version="1.0">
+    <option name="myName" value="Project Default" />
+    <inspection_tool class="PyPackageRequirementsInspection" enabled="true" level="WARNING" enabled_by_default="true">
+      <option name="ignoredPackages">
+        <value>
+          <list size="1">
+            <item index="0" class="java.lang.String" itemvalue="pytest-runner" />
+          </list>
+        </value>
+      </option>
+    </inspection_tool>
+    <inspection_tool class="PyPep8NamingInspection" enabled="true" level="WEAK WARNING" enabled_by_default="true">
+      <option name="ignoredErrors">
+        <list>
+          <option value="N801" />
+          <option value="N806" />
+        </list>
+      </option>
+    </inspection_tool>
+    <inspection_tool class="PyUnresolvedReferencesInspection" enabled="true" level="WARNING" enabled_by_default="true">
+      <option name="ignoredIdentifiers">
+        <list>
+          <option value="dict.labels" />
+          <option value="geograpy.places.countries" />
+        </list>
+      </option>
+    </inspection_tool>
+  </profile>
+</component>

.idea/inspectionProfiles/profiles_settings.xml

@@ -0,0 +1,6 @@
+<component name="InspectionProjectProfileManager">
+  <settings>
+    <option name="USE_PROJECT_PROFILE" value="false" />
+    <version value="1.0" />
+  </settings>
+</component>

.idea/misc.xml

@@ -0,0 +1,4 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="ProjectRootManager" version="2" project-jdk-name="Python 3.10 (sandbox)" project-jdk-type="Python SDK" />
+</project>

.idea/model-card-regulatory-check.iml

@@ -0,0 +1,8 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<module type="PYTHON_MODULE" version="4">
+  <component name="NewModuleRootManager">
+    <content url="file://$MODULE_DIR$" />
+    <orderEntry type="jdk" jdkName="Python 3.10 (sandbox)" jdkType="Python SDK" />
+    <orderEntry type="sourceFolder" forTests="false" />
+  </component>
+</module>

.idea/modules.xml

@@ -0,0 +1,8 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="ProjectModuleManager">
+    <modules>
+      <module fileurl="file://$PROJECT_DIR$/.idea/model-card-regulatory-check.iml" filepath="$PROJECT_DIR$/.idea/model-card-regulatory-check.iml" />
+    </modules>
+  </component>
+</project>

.idea/vcs.xml

@@ -0,0 +1,6 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="VcsDirectoryMappings">
+    <mapping directory="" vcs="Git" />
+  </component>
+</project>

Dockerfile

@@ -0,0 +1,27 @@
+FROM python:3.11-slim-bullseye
+
+# Set the working directory to /code
+WORKDIR /code
+
+# Copy the current directory contents into the container at /code
+COPY ./requirements.txt /code/requirements.txt
+
+# Install requirements.txt
+RUN pip install --no-cache-dir --upgrade -r /code/requirements.txt
+
+# Set up a new user named "user" with user ID 1000
+RUN useradd -m -u 1000 user
+# Switch to the "user" user
+USER user
+# Set home to the user's home directory
+ENV HOME=/home/user \
+	PATH=/home/user/.local/bin:$PATH
+
+# Set the working directory to the user's home directory
+WORKDIR $HOME/app
+
+# Copy the current directory contents into the container at $HOME/app setting the owner to the user
+COPY --chown=user . $HOME/app
+
+EXPOSE 7860
+CMD ["uvicorn", "server:app","--proxy-headers", "--host", "0.0.0.0", "--port", "7860"]

main.py

@@ -0,0 +1,154 @@
+import os
+from typing import Dict, Any, Optional, List
+import re
+from abc import ABC, abstractmethod
+
+from huggingface_hub import (ModelCard, comment_discussion,
+                             create_discussion, get_discussion_details,
+                             get_repo_discussions)
+import markdown
+from bs4 import BeautifulSoup
+from tabulate import tabulate
+from difflib import SequenceMatcher
+
+KEY = os.environ.get("KEY")
+
+
+def similar(a, b):
+    """Check similarity of two sequences"""
+    return SequenceMatcher(None, a, b).ratio()
+
+
+class ComplianceCheck(ABC):
+    def __init__(self, name):
+        self.name = name
+
+    @abstractmethod
+    def check(self, card: BeautifulSoup) -> bool:
+        raise NotImplementedError
+
+
+class ModelProviderIdentityCheck(ComplianceCheck):
+    def __init__(self):
+        super().__init__("Identity and Contact Details")
+
+    def check(self, card: BeautifulSoup):
+        developed_by_li = card.findAll(text=re.compile("Developed by"))[0].parent.parent
+        developed_by = list(developed_by_li.children)[1].text.strip()
+
+        if developed_by == "[More Information Needed]":
+            return False
+        else:
+            return True
+
+
+class IntendedPurposeCheck(ComplianceCheck):
+    def __init__(self):
+        super().__init__("Intended Purpose")
+
+    def check(self, card: BeautifulSoup):
+
+        # direct_use = card.find_all("h2", text="Direct Use")[0]
+        #
+        # if developed_by == "[More Information Needed]":
+        #     return False
+        # else:
+        return False
+
+
+compliance_checks = [
+    ModelProviderIdentityCheck(),
+    IntendedPurposeCheck()
+    # "General Limitations",
+    # "Computational and Hardware Requirements",
+    # "Carbon Emissions"
+]
+
+
+def parse_webhook_post(data: Dict[str, Any]) -> Optional[str]:
+    event = data["event"]
+    if event["scope"] != "repo":
+        return None
+    repo = data["repo"]
+    repo_name = repo["name"]
+    repo_type = repo["type"]
+    if repo_type != "model":
+        raise ValueError("Incorrect repo type.")
+    return repo_name
+
+
+def check_compliance(comp_checks: List[ComplianceCheck], card: BeautifulSoup) -> Dict[str, bool]:
+    return {c.name: c.check(card) for c in comp_checks}
+
+
+def run_compliance_check(repo_name):
+    card_data: ModelCard = ModelCard.load(repo_id_or_path=repo_name)
+    card_html = markdown.markdown(card_data.content)
+    card_soup = BeautifulSoup(card_html, features="html.parser")
+    compliance_results = check_compliance(compliance_checks, card_soup)
+
+    return compliance_results
+
+
+def create_metadata_breakdown_table(compliance_check_dictionary):
+    data = {k: v for k, v in compliance_check_dictionary.items()}
+    metadata_fields_column = list(data.keys())
+    metadata_values_column = list(data.values())
+    table_data = list(zip(metadata_fields_column, metadata_values_column))
+    return tabulate(
+        table_data, tablefmt="github", headers=("Compliance Check", "Present")
+    )
+
+
+def create_markdown_report(
+    desired_metadata_dictionary, repo_name, update: bool = False
+):
+    report = f"""# Model Card Regulatory Compliance report card {"(updated)" if update else ""}
+    \n
+This is an automatically produced model card regulatory compliance report card for {repo_name}.
+This report is meant as a POC!
+    \n 
+## Breakdown of metadata fields for your model
+\n
+{create_metadata_breakdown_table(desired_metadata_dictionary)}
+\n
+    """
+    return report
+
+
+def create_or_update_report(compliance_check, repo_name):
+    report = create_markdown_report(
+        compliance_check, repo_name, update=False
+    )
+    repo_discussions = get_repo_discussions(
+        repo_name,
+        repo_type="model",
+    )
+    for discussion in repo_discussions:
+        if (
+            discussion.title == "Metadata Report Card" and discussion.status == "open"
+        ):  # An existing open report card thread
+            discussion_details = get_discussion_details(
+                repo_name, discussion.num, repo_type="model"
+            )
+            last_comment = discussion_details.events[-1].content
+            if similar(report, last_comment) <= 0.999:
+                report = create_markdown_report(
+                    compliance_check,
+                    repo_name,
+                    update=True,
+                )
+                comment_discussion(
+                    repo_name,
+                    discussion.num,
+                    comment=report,
+                    repo_type="model",
+                )
+            return True
+    create_discussion(
+        repo_name,
+        "Model Card Regulatory Compliance Report Card",
+        description=report,
+        repo_type="model",
+    )
+    return True

requirements.txt

@@ -0,0 +1,5 @@
+fastapi
+uvicorn
+markdown
+beautifulsoup4
+tabulate

server.py

@@ -0,0 +1,25 @@
+import os
+from fastapi import FastAPI, Request, Response
+from main import parse_webhook_post, run_compliance_check, create_or_update_report
+
+KEY = os.environ.get("KEY")
+
+app = FastAPI()
+
+
+@app.post("/webhook")
+async def webhook(request: Request):
+    if request.method == "POST":
+        # if request.headers.get("X-Webhook-Secret") != KEY:
+        #     return Response("Invalid secret", status_code=401)
+
+        data = await request.json()
+
+        if parsed_post := parse_webhook_post(data):
+            repo_name = parsed_post
+        else:
+            return Response("Unable to parse webhook data", status_code=400)
+
+        compliance_check = run_compliance_check(repo_name)
+        result = create_or_update_report(compliance_check, repo_name)
+        return result