Saving local versions of Clip-Engine core files

.gitattributes

@@ -0,0 +1,35 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,12 @@
+---
+title: Clip Engine
+emoji: ⚡
+colorFrom: yellow
+colorTo: indigo
+sdk: gradio
+sdk_version: 5.33.2
+app_file: app.py
+pinned: false
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,133 @@
+import gradio as gr
+from pytube import YouTube
+from textblob import TextBlob
+from transformers import pipeline
+import logging
+import os
+
+# Configure basic logging
+logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
+
+# Initialize sentiment analysis pipeline (once, to save resources)
+# Using a specific model for potentially better results or if default is too large/slow
+# You might need to adjust the model based on availability and performance on HF Spaces free tier
+SENTIMENT_MODEL_NAME = "distilbert-base-uncased-finetuned-sst-2-english"
+sentiment_analyzer = None
+try:
+    logging.info(f"Attempting to load sentiment analysis pipeline: {SENTIMENT_MODEL_NAME}...")
+    # Specify a cache directory within the Space's writable area if needed
+    # cache_dir = "/data/.cache/huggingface/transformers" # Example for some HF environments
+    # os.makedirs(cache_dir, exist_ok=True)
+    sentiment_analyzer = pipeline("sentiment-analysis", model=SENTIMENT_MODEL_NAME) #, cache_dir=cache_dir)
+    logging.info(f"Sentiment analysis pipeline '{SENTIMENT_MODEL_NAME}' loaded successfully.")
+except Exception as e:
+    logging.warning(f"Failed to load {SENTIMENT_MODEL_NAME}: {e}. Falling back to default sentiment model.")
+    try:
+        sentiment_analyzer = pipeline("sentiment-analysis") #, cache_dir=cache_dir)
+        logging.info("Default sentiment analysis pipeline loaded successfully.")
+    except Exception as e_default:
+        logging.error(f"Failed to load default sentiment analysis pipeline: {e_default}")
+        # sentiment_analyzer will remain None
+
+
+def analyze_youtube_content(youtube_url: str = "", transcript_text: str = "") -> dict:
+    """
+    Analyzes YouTube video content.
+    If a YouTube URL is provided, it attempts to fetch video information (e.g., title, views, length).
+    If transcript text is provided, it performs sentiment analysis on the text using
+    TextBlob and a Hugging Face transformer model.
+
+    Args:
+        youtube_url (str, optional): The URL of the YouTube video. Defaults to "".
+        transcript_text (str, optional): The transcript text of the video. Defaults to "".
+
+    Returns:
+        dict: A dictionary containing analysis results.
+              Includes 'video_info' if URL is processed and 'sentiment_analysis' if transcript is processed.
+    """
+    results = {}
+    logging.info(f"Tool 'analyze_youtube_content' called with URL: '{youtube_url}', Transcript provided: {bool(transcript_text)}")
+
+    if not youtube_url and not transcript_text:
+        logging.warning("No YouTube URL or transcript text provided.")
+        return {"error": "No YouTube URL or transcript text provided for analysis."}
+
+    if youtube_url:
+        try:
+            yt = YouTube(youtube_url)
+            results["video_info"] = {
+                "title": yt.title,
+                "views": yt.views,
+                "length_seconds": yt.length,
+                "publish_date": yt.publish_date.strftime('%Y-%m-%d') if yt.publish_date else None,
+                "status": "success"
+            }
+            logging.info(f"Successfully fetched info for video: {yt.title}")
+        except Exception as e:
+            logging.error(f"Error fetching video info from URL '{youtube_url}': {e}")
+            results["video_info"] = {
+                "status": "error",
+                "message": f"Could not fetch video info: {str(e)}"
+            }
+
+    if transcript_text:
+        analysis_data = {}
+        # TextBlob sentiment
+        try:
+            blob = TextBlob(transcript_text)
+            tb_sentiment = blob.sentiment
+            analysis_data["textblob"] = {
+                "polarity": round(tb_sentiment.polarity, 3),
+                "subjectivity": round(tb_sentiment.subjectivity, 3),
+                "assessment": "positive" if tb_sentiment.polarity > 0.05 else "negative" if tb_sentiment.polarity < -0.05 else "neutral"
+            }
+            logging.info("TextBlob sentiment analysis complete.")
+        except Exception as e:
+            logging.error(f"Error during TextBlob sentiment analysis: {e}")
+            analysis_data["textblob"] = {"error": str(e)}
+
+        # Hugging Face sentiment
+        if sentiment_analyzer:
+            try:
+                # Truncate for performance and model limits (default for many models is 512 tokens)
+                max_length_chars = 1000 # Heuristic, actual token limit is what matters
+                truncated_text = transcript_text[:max_length_chars]
+                hf_result = sentiment_analyzer(truncated_text)[0]
+                analysis_data["huggingface_transformer"] = {
+                    "label": hf_result["label"],
+                    "score": round(hf_result["score"], 3)
+                }
+                if len(transcript_text) > max_length_chars:
+                    analysis_data["huggingface_transformer"]["note"] = f"Analyzed approximately the first {max_length_chars} characters of the transcript."
+                logging.info("Hugging Face transformer sentiment analysis complete.")
+            except Exception as e:
+                logging.error(f"Error during Hugging Face sentiment analysis: {e}")
+                analysis_data["huggingface_transformer"] = {"error": str(e)}
+        else:
+            analysis_data["huggingface_transformer"] = {"error": "Hugging Face sentiment analyzer not loaded."}
+            logging.warning("Hugging Face sentiment analyzer was not available for analysis.")
+
+        results["sentiment_analysis"] = analysis_data
+
+    if not results:
+         return {"status": "No analysis performed, though input was provided. Check logs."}
+
+    return results
+
+# Create Gradio interface for the tool.
+youtube_tool_interface = gr.Interface(
+    fn=analyze_youtube_content,
+    inputs=[
+        gr.Textbox(label="YouTube Video URL (Optional)", placeholder="Enter YouTube video URL..."),
+        gr.Textbox(label="Video Transcript Text (Optional)", placeholder="Paste video transcript here...", lines=5)
+    ],
+    outputs=gr.JSON(label="Analysis Result"),
+    title="YouTube Content Analyzer Tool",
+    description="Provides information and sentiment analysis for a YouTube video URL or its transcript. (For Agent Use via MCP)"
+)
+
+# Launch the Gradio app with the MCP server enabled.
+if __name__ == "__main__":
+    logging.info("Launching Gradio app with MCP server enabled for the YouTube Content Analyzer Tool...")
+    # The `mcp_server=True` flag is crucial for the agent to connect and use the tool.
+    youtube_tool_interface.launch(mcp_server=True)