Update generate_readme.py

Author: Code-master31

.github/scripts/generate_readme.py

@@ -1,35 +1,113 @@
+#!/usr/bin/env python3
 import os
+import json
 
-BASE_DIR = "."
-TEMPLATE = """# {project_name}
+# If you keep projects in a "projects" folder change this variable to 'projects'
+PROJECTS_ROOT = "projects" if os.path.isdir("projects") else "."
+
+EXCLUDE_FOLDERS = {".github", ".git", ".vscode", "node_modules"}
+
+README_TEMPLATE = """# {title}
 
 ## 📌 Project Overview
-This project is part of the **Machine Learning Projects Collection (100+ projects with source code)**.  
+{description}
 
 ## 🛠️ Potential Technologies / Tools
-- Python
-- Pandas
-- Scikit-learn
-- TensorFlow/Keras (if deep learning)
+{tools_list}
 
 ## 🚀 How to Run
 1. Clone this repository
 2. Navigate to this project folder
-3. Install dependencies (`pip install -r requirements.txt`)
-4. Run the code  
+3. Install dependencies (e.g. `pip install -r requirements.txt`)
+4. Run the code (follow project-specific instructions)
 
 ---
-✅ Auto-generated by [ML Projects Bot] 🤖
+✅ Auto-generated by ML Projects Bot 🤖
 """
 
-def format_name(folder):
-    return folder.replace("-", " ").replace("_", " ").title()
-
-for folder in os.listdir(BASE_DIR):
-    if os.path.isdir(folder) and folder not in [".github", ".git"]:
-        readme_path = os.path.join(folder, "README.md")
-        if not os.path.exists(readme_path):
-            project_name = format_name(folder)
-            with open(readme_path, "w") as f:
-                f.write(TEMPLATE.format(project_name=project_name))
-            print(f"Generated README for {project_name}")
+DEFAULT_DESCRIPTION = "This project is part of the Machine-learning-projects-collection-100-plus-projects-with-source-code repository."
+DEFAULT_TOOLS = "Python, Pandas, Scikit-learn (customize for your project)"
+
+KEYWORD_STACKS = {
+    "spam": "Python, Scikit-learn, NLTK, Naive Bayes",
+    "sentiment": "Python, NLTK, Scikit-learn, TextBlob",
+    "stock": "Pandas, yfinance, LSTM, TensorFlow/Keras",
+    "price": "Pandas, Scikit-learn, Regression / Time-series",
+    "image": "TensorFlow/Keras, OpenCV, CNN",
+    "ocr": "Tesseract, OpenCV, EasyOCR",
+    "fraud": "Scikit-learn, IsolationForest, XGBoost",
+    "classification": "Scikit-learn, RandomForest, LogisticRegression",
+    "regression": "Scikit-learn, LinearRegression, XGBoost",
+    "segmentation": "U-Net, TensorFlow/PyTorch",
+}
+
+def format_title(name):
+    return name.replace("-", " ").replace("_", " ").title()
+
+def guess_tools(name):
+    lower = name.lower()
+    for k,v in KEYWORD_STACKS.items():
+        if k in lower:
+            return v
+    return DEFAULT_TOOLS
+
+def load_metadata(folder):
+    # Supports optional project.json containing {"name": "...", "description": "...", "tools": ["t1","t2"]}
+    meta_path = os.path.join(folder, "project.json")
+    if os.path.exists(meta_path):
+        try:
+            with open(meta_path, "r", encoding="utf-8") as f:
+                return json.load(f)
+        except Exception as e:
+            print(f"Could not parse {meta_path}: {e}")
+    return None
+
+def write_readme(folder, title, description, tools):
+    readme_path = os.path.join(folder, "README.md")
+    if os.path.exists(readme_path):
+        print(f"README exists for {folder}, skipping.")
+        return False
+    tools_list = "\n".join(f"- {t}" for t in (tools.split(",") if isinstance(tools, str) else tools))
+    content = README_TEMPLATE.format(title=title, description=description, tools_list=tools_list)
+    with open(readme_path, "w", encoding="utf-8") as f:
+        f.write(content)
+    print(f"Generated README.md for {folder}")
+    return True
+
+def main():
+    base = os.getcwd()
+    candidates = []
+    if PROJECTS_ROOT == ".":
+        # all top-level directories except excluded ones
+        for name in os.listdir("."):
+            if os.path.isdir(name) and name not in EXCLUDE_FOLDERS:
+                candidates.append(name)
+    else:
+        root_path = os.path.join(base, PROJECTS_ROOT)
+        for name in os.listdir(root_path):
+            full = os.path.join(root_path, name)
+            if os.path.isdir(full) and name not in EXCLUDE_FOLDERS:
+                candidates.append(full)
+
+    any_written = False
+    for folder in sorted(candidates):
+        full_path = folder if PROJECTS_ROOT == "." else os.path.join(PROJECTS_ROOT, os.path.basename(folder))
+        meta = load_metadata(full_path)
+        if meta:
+            title = meta.get("name", format_title(os.path.basename(full_path)))
+            description = meta.get("description", DEFAULT_DESCRIPTION)
+            tools = meta.get("tools", guess_tools(os.path.basename(full_path)))
+            tools_str = ", ".join(tools) if isinstance(tools, list) else tools
+        else:
+            title = format_title(os.path.basename(full_path))
+            description = DEFAULT_DESCRIPTION
+            tools_str = guess_tools(os.path.basename(full_path))
+
+        created = write_readme(full_path, title, description, tools_str)
+        any_written = any_written or created
+
+    if not any_written:
+        print("No new README files generated.")
+
+if __name__ == "__main__":
+    main()