Add multimodality table to documentation

docs/generate_model_js.py

@@ -1,5 +1,6 @@
 import json
-
+import os
+import glob
 
 def get_key_val(part):
     key_index_start = part.index('[')
@@ -13,48 +14,8 @@ def get_key_val(part):
     return key, val
 
 
-# Read the content from README.md
-with open('../README.md', 'r') as file:
-    content = file.read()
-
-# Extract the relevant information from the content
-models = []
-lines = content.split('\n')
-lines = lines[lines.index('## Models') + 4: lines.index('## Resources') - 2]
-
-headers = []
-headers = lines[0].split('|')[1:-1]
-headers = [header.strip() for header in headers]
-
-for line in lines[2:]:
-    parts = line.split('|')[1:-1]
-    parts = [part.strip() for part in parts]
-    model = dict(zip(headers, parts))
-    models.append(model)
-
-year = None
-
-for model in models:
-    # handle empty years
-    if model["Year"] == "":
-        model["Year"] = year
-    else:
-        year = model["Year"]
-
-    # handle model, docs and paper part
-    name_paper_str = model["Model and Paper"]
-
-    for i, part in enumerate(name_paper_str.split(', ')):
-        key, val = get_key_val(part)
-
-        if i == 0:
-            model["Name"] = key
-            model["Link"] = val
-        else:
-            model[key] = val
-
+def add_pytorch_tensorflow(model):
     # handle environment part
-
     env_part = model["Environment"].split(', ')[0]
 
     search_dict = {
@@ -74,28 +35,104 @@ def get_key_val(part):
     else:
         for header, _ in search_dict.items():
             model[header] = False
+
+
+def add_modalities_for_model(model):
+    modalities_keywords = {
+        "User Text": "user_text",
+        "User Image": "user_image",
+        "User Graph": "user_graph",
+        "Item Text": "item_text",
+        "Item Image": "item_image",
+        "Item Graph": "item_graph",
+        "Sentiment": "sentiment",
+        "Review Text": "review_text",
+        "User Feature": "user_feature",
+        "Item Feature": "item_feature",
+    }
 
-    # remove non required keys
-    model.pop("Model and Paper")
-    model.pop("Environment")
-
-    # Get package name
-    model_dir = model["Link"]
+    for filename in glob.glob(f'../{model["Link"]}/*.py', recursive=True):
+        with open(filename, 'r') as file:
+            file_data = file.read()
+
+            for header, modality_keyword in modalities_keywords.items():
+                is_found = modality_keyword in file_data
+                if is_found:
+                    model[header] = True
 
-    with open(f'../{model_dir}/__init__.py', 'r') as file:
-        init_data = file.read()
+    for header, modality_keyword in modalities_keywords.items():
+        if header not in model:
+            model[header] = False
+
+
+if __name__ == "__main__":
+    # Read the content from README.md
+    with open('../README.md', 'r') as file:
+        content = file.read()
+
+    # Extract the relevant information from the content
+    models = []
+    lines = content.split('\n')
+    lines = lines[lines.index('## Models') + 4: lines.index('## Resources') - 2]
+
+    headers = []
+    headers = lines[0].split('|')[1:-1]
+    headers = [header.strip() for header in headers]
+
+    for line in lines[2:]:
+        parts = line.split('|')[1:-1]
+        parts = [part.strip() for part in parts]
+        model = dict(zip(headers, parts))
+        models.append(model)
+
+    year = None
+
+    for model in models:
+        # handle empty years
+        if model["Year"] == "":
+            model["Year"] = year
+        else:
+            year = model["Year"]
+
+        # handle model, docs and paper part
+        name_paper_str = model["Model and Paper"]
+
+        for i, part in enumerate(name_paper_str.split(', ')):
+            key, val = get_key_val(part)
+
+            if i == 0:
+                model["Name"] = key
+                model["Link"] = val
+            else:
+                model[key] = val
 
-        package_names = []
+        # Check for PyTorch and TensorFlow in each requirements file
+        add_pytorch_tensorflow(model)
 
-        for row in init_data.split('\n'):
-            if "import" in row:
-                package_name = row[row.index("import") + len("import "):]
-                package_names.append(f"cornac.models.{package_name}")
+        # Check for modalities keywords in files and add to model
+        add_modalities_for_model(model)
+
+        # remove non required keys
+        model.pop("Model and Paper")
+        model.pop("Environment")
 
-        model["packages"] = package_names
+        # Get package name
+        model_dir = model["Link"]
+
+        with open(f'../{model_dir}/__init__.py', 'r') as file:
+            init_data = file.read()
+
+            package_names = []
+
+            for row in init_data.split('\n'):
+                if "import" in row:
+                    package_name = row[row.index("import") + len("import "):]
+                    package_names.append(f"cornac.models.{package_name}")
+
+            model["packages"] = package_names
 
-json_str = json.dumps(models, indent=4)
+    json_str = json.dumps(models, indent=4)
 
-# Write the JSON object to a file
-with open('source/_static/models/data.js', 'w') as file:
-    file.write(f"var data = {json_str};")
+    # Write the JSON object to a file
+    with open('source/_static/models/data.js', 'w') as file:
+        file.write(f"var data = {json_str};")