Upload folder using huggingface_hub

README.md

@@ -1,12 +1,6 @@
 ---
-title: Brighter India Yt Scraper
-emoji: 🐢
-colorFrom: red
-colorTo: red
+title: brighter-india-yt-scraper
+app_file: scraper.py
 sdk: gradio
 sdk_version: 4.19.2
-app_file: app.py
-pinned: false
 ---
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

requirements.txt

@@ -0,0 +1,4 @@
+gradio
+pandas
+transformers
+google-api-python-client

scraper.py

@@ -0,0 +1,69 @@
+import gradio as gr
+import os
+import pandas as pd
+from googleapiclient.discovery import build
+from transformers import pipeline
+
+# Set up YouTube Data API credentials and initialize
+api_key = "AIzaSyBUX6ak7fd2KEh-2aUM_aH26jVEw6Wj5V4"  # Replace with your own API key
+youtube = build('youtube', 'v3', developerKey=api_key)
+
+# Initialize sentiment analysis pipeline
+sentiment_pipeline = pipeline("sentiment-analysis")
+
+def get_video_comments(video_id, max_length=512):
+    comments = []
+    next_page_token = None
+    
+    while True:
+        response = youtube.commentThreads().list(
+            part='snippet',
+            videoId=video_id,
+            pageToken=next_page_token if next_page_token else ''
+        ).execute()
+
+        for item in response['items']:
+            comment = item['snippet']['topLevelComment']['snippet']['textDisplay']
+            # Truncate the comment if it exceeds the maximum length
+            comment = comment[:max_length]
+            comments.append(comment)
+
+        next_page_token = response.get('nextPageToken')
+
+        if not next_page_token:
+            break
+
+    return comments
+
+def analyze_sentiment(comments):
+    if comments:  # Ensure there are comments to analyze
+        results = sentiment_pipeline(comments)
+        return results
+    else:
+        return []
+
+def process_video(yt_link):
+    video_id = yt_link.split("=")[-1]  # Extract video ID from the link
+    comments = get_video_comments(video_id)
+    sentiment_results = analyze_sentiment(comments)
+
+    # Create a DataFrame from the comments and sentiment analysis results
+    df = pd.DataFrame({
+        'Comments': comments,
+        'Sentiment': [result['label'] for result in sentiment_results],
+        'Score': [result['score'] for result in sentiment_results]
+    })
+
+    return df
+
+# Define the Gradio interface
+iface = gr.Interface(
+    fn=process_video,
+    inputs=gr.Textbox(lines=2, placeholder="Enter YouTube video URL here..."),
+    outputs="dataframe",
+    title="YouTube Video Comments Sentiment Analysis",
+    description="Enter a YouTube video link to analyze the sentiment of its comments."
+)
+
+# Launch the interface
+iface.launch(share=True)

youtube_comments_sentiment.csv

@@ -0,0 +1,19 @@
+Comments,Sentiment,Score
+❤❤❤❤❤❤❤❤,NEGATIVE,0.6970567107200623
+"Che bellezza, è un amore❤❤❤❤",POSITIVE,0.7507229447364807
+couch insurance (what’s a scratching post ?),NEGATIVE,0.998101532459259
+"Forps, you can bite my hand anytime you want to.  Okay?  Love you, Forps.",POSITIVE,0.9997705817222595
+The insurance will be void because you actually have a squirrel.,NEGATIVE,0.9991538524627686
+<Announcer Voice> Forps wins!,POSITIVE,0.9996067881584167
+Reason for insurance: Forps the squirrel-raccoon!,NEGATIVE,0.975786566734314
+Forps with the wide-eyed stare!,POSITIVE,0.9920353293418884
+That love bite  ❤,POSITIVE,0.9913235902786255
+Forps is fierce.,POSITIVE,0.9994118213653564
+😍🥰😘🐾🐈‍⬛🥰😍forps,NEGATIVE,0.6970567107200623
+"Forps is too big, this is scary more than cute now 😮 still very cute 😅",POSITIVE,0.946954071521759
+Flooficus Majesticus,POSITIVE,0.9862658381462097
+😂👍,NEGATIVE,0.6970567107200623
+schöne Katze,NEGATIVE,0.9197412133216858
+naughty:3,NEGATIVE,0.6425185203552246
+The last stop punch tho 😅,NEGATIVE,0.9785932302474976
+"Sassy little guy, isn’t he?",POSITIVE,0.9434502720832825