initial commit

Dockerfile

@@ -0,0 +1,156 @@
+FROM python:3.9
+WORKDIR /app
+COPY requirements.txt /app/requirements.txt
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Use the recommended HF_HOME instead of deprecated TRANSFORMERS_CACHE
+ENV HF_HOME=/code/cache/huggingface
+
+# Create the directory for the Transformers cache and set permissions
+RUN mkdir -p /code/cache/huggingface && chmod -R 777 /code/cache/huggingface
+
+COPY . /app
+
+EXPOSE 7860
+# Increase Gunicorn timeout to prevent worker timeout during long initializations
+CMD ["gunicorn", "-b", "0.0.0.0:7860", "main:app", "--timeout", "120", "--workers", "2", "--threads", "2"]
+
+
+# # Use an official Python runtime as a base image
+# FROM python:3.9
+
+# # Set the working directory in the container
+# WORKDIR /app
+
+# # Copy the requirements file into the container at /app
+# COPY ./requirements.txt /code/requirements.txt
+
+# # Install any needed packages specified in requirements.txt
+# RUN pip install --no-cache-dir --upgrade -r /code/requirements.txt
+
+# # Define environment variable for the Hugging Face home
+# # ENV HF_HOME=/app/cache/huggingface
+# ENV HF_HOME=/code/cache/huggingface
+
+
+# # Create the directory for the Transformers cache and set permissions
+# RUN mkdir -p /code/cache/huggingface && chmod -R 777 /code/cache/huggingface
+
+
+# # # Create the directory for the Hugging Face cache
+# # RUN mkdir -p $HF_HOME
+# # Optional: Adjust permissions if necessary
+# # RUN chmod 755 $HF_HOME
+
+# # Copy the rest of your application's code into the container at /app
+# # COPY . /app
+# COPY . .
+
+# # Define environment variable for the Flask application port
+# # ENV PORT=8080
+
+# # # Expose the port the application runs on
+# # EXPOSE 7860
+# CMD ["panel", "server", "/code/app/py", "--address","0.0.0.0","--port", "7860" "--allow-websocket-origin","pankaj100567-Textsimilarity-str`"]
+
+# CMD ["gunicorn", "-b", "0.0.0.0:7860", "main:app"]
+
+# Run app.py when the container launches
+# CMD ["python", "app.py", "--allow-websocket-origin","pankaj100567-Textsimilarity-str.hf.space"]
+# CMD ["python", "app.py", "--allow-websocket-origin","pankaj100567-Textsimilarity-str.hf.space"]
+
+
+
+# # Use an official Python runtime as a base image
+# FROM python:3.9
+
+# # Set the working directory in the container
+# WORKDIR /app
+
+# # Copy the requirements file into the container at /app
+# COPY requirements.txt /app/requirements.txt
+
+# # Install any needed packages specified in requirements.txt
+# RUN pip install --no-cache-dir -r requirements.txt
+
+# # Define environment variable for the Transformers cache
+# ENV TRANSFORMERS_CACHE=/app/cache/huggingface
+
+# # Create the directory for the Transformers cache
+# RUN mkdir -p /app/cache/huggingface && chmod 777 /app/cache/huggingface
+
+# # Copy the rest of your application's code into the container at /app
+# COPY . /app
+
+# # Define environment variable for the Flask application port
+# ENV PORT=8080
+
+# # Expose the port the application runs on
+# EXPOSE 8080
+
+# # Run app.py when the container launches
+# CMD ["python", "app.py"]
+
+
+# # Use an official Python runtime as a base image
+# FROM python:3.9
+
+# # Set the working directory in the container
+# WORKDIR /app
+
+# # Copy the requirements file into the container at /app
+# COPY requirements.txt /app/requirements.txt
+
+# # Install any needed packages specified in requirements.txt
+# RUN pip install --no-cache-dir -r requirements.txt
+
+# # Copy the rest of your application's code into the container at /app
+# COPY . /app
+
+# # Define environment variable
+# ENV PORT 8080
+
+# # Run app.py when the container launches
+# CMD ["python", "app.py"]
+
+
+# FROM python:3.9
+
+
+# WORKDIR /code
+
+# COPY ./requirements.txt /code/requirements.txt
+
+# RUN pip install --no-cache-dir --upgrade -r /code/requirements.txt
+
+# COPY . .
+
+# CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "7860"]
+
+# # Use an official Python runtime as a parent image
+# FROM python:3.9
+
+# # Set the working directory in the container
+# WORKDIR /code
+
+# # Copy the dependencies file to the working directory
+# COPY requirements.txt /code/
+
+# # Install any needed packages specified in requirements.txt
+# RUN pip install --no-cache-dir --upgrade -r requirements.txt
+
+# # Copy the current directory contents into the container at /code
+# COPY . /code/
+
+# # Make port 5000 available to the world outside this container
+# EXPOSE 5000
+
+# # Define environment variable
+# ENV FLASK_APP=app.py
+# ENV FLASK_RUN_HOST=0.0.0.0
+# ENV FLASK_RUN_PORT=5000
+
+# # Run the application when the container launches
+# # CMD ["flask", "run"]
+
+# CMD ["flask","run","panel","server","--allow-websocker-origin","pankaj100567-similarity-measure.hf.space"]

main.py

@@ -0,0 +1,74 @@
+from flask import Flask, jsonify, request, render_template
+import torch
+# from langdetect import detect, DetectorFactory
+
+import os
+from transformers import XLMRobertaForSequenceClassification, AutoTokenizer
+from transformers import RobertaForSequenceClassification
+
+app = Flask(__name__)
+
+class Predictor:
+    def __init__(self, model, tokenizer, device):
+        self.model = model
+        self.tokenizer = tokenizer
+        self.device = device
+
+    def predict_similarity(self, sentence1, sentence2):
+        try:
+            # Tokenize input sentences
+            encoded_input = self.tokenizer(sentence1, sentence2, return_tensors='pt', padding=True, truncation=True)
+            input_ids = encoded_input['input_ids'].to(self.device)
+            attention_mask = encoded_input['attention_mask'].to(self.device)
+
+            # Perform inference
+            with torch.no_grad():
+                outputs = self.model(input_ids=input_ids, attention_mask=attention_mask)
+                logits = outputs.logits
+                similarity_score = torch.sigmoid(logits).item()  # Assuming binary classification
+            return similarity_score
+        except Exception as e:
+            print(f"Error during model prediction: {e}")
+            return 0.0  # Return a default or error value if any exception occurs
+
+# Load model and tokenizer
+# model_path = "pankaj100567/semantic_textual_relatedness"
+# model_path="pankaj100567/str_english_model_roberta_large_1stage"
+# model_path= "epoch_1"
+# model_path="pankaj100567/semantic-english-model"
+model_path="pankaj100567/semeval-semantic-texutal-relatedness"
+# cache_dir = "/app/cache/huggingface"
+cache_dir = "/code/cache/huggingface"
+if not os.path.exists(cache_dir):
+    try:
+        os.makedirs(cache_dir)
+        os.chmod(cache_dir, 0o777)  # Set directory permissions to read, write, and execute by all users
+    except Exception as e:
+        print(f"Failed to create or set permissions for directory {cache_dir}: {e}")
+
+model = XLMRobertaForSequenceClassification.from_pretrained(model_path, cache_dir= cache_dir, num_labels=1)
+tokenizer = AutoTokenizer.from_pretrained("xlm-roberta-large",cache_dir= cache_dir,)
+
+# model = XLMRobertaForSequenceClassification.from_pretrained(model_path)
+# tokenizer = AutoTokenizer.from_pretrained("xlm-roberta-large")
+
+# Device configuration
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+model.to(device)
+
+# Initialize Predictor instance
+predictor = Predictor(model, tokenizer, device)
+
+@app.route('/')
+def index():
+    return render_template('index.html')
+
+@app.route('/predict', methods=['POST'])
+def predict():
+    sentence1 = request.form['sentence1']
+    sentence2 = request.form['sentence2']
+    similarity_score = predictor.predict_similarity(sentence1, sentence2)
+    return render_template('result.html', sentence1=sentence1, sentence2=sentence2, similarity_score=similarity_score)
+
+# if __name__ == '__main__':
+#     app.run(debug=True, host='0.0.0.0', port=5002)  # Ensure the app is accessible externally

requirements.txt

@@ -0,0 +1,25 @@
+gunicorn
+transformers
+accelerate
+einops
+xformers
+bitsandbytes
+huggingface_hub
+pypdf
+torch
+datasets
+numpy
+scipy
+flask
+gunicorn
+jinja2
+pandas
+matplotlib
+tokenizers
+scikit-learn
+gradio
+gunicorn
+nltk
+langdetect
+beautifulsoup4
+Django

static/styles.css

@@ -0,0 +1,57 @@
+/* styles.css */
+
+body {
+    font-family: Arial, sans-serif;
+    background-color: #d13030;
+    margin: 0;
+    padding: 0;
+}
+
+.container {
+    width: 80%;
+    margin: 50px auto;
+    background-color: #310c0c;
+    padding: 20px;
+    border-radius: 10px;
+    box-shadow: 0px 0px 10px rgba(0, 0, 0, 0.1);
+}
+
+h1 {
+    text-align: center;
+    color: #333;
+}
+
+.input-group {
+    margin-bottom: 20px;
+}
+
+.input-group label {
+    display: block;
+    margin-bottom: 5px;
+    font-weight: bold;
+}
+
+.input-group input[type="text"] {
+    width: 100%;
+    padding: 10px;
+    font-size: 16px;
+    border: 1px solid #ccc;
+    border-radius: 5px;
+}
+
+button {
+    display: block;
+    width: 100%;
+    padding: 10px;
+    font-size: 16px;
+    background-color: #007bff;
+    color: #fff;
+    border: none;
+    border-radius: 5px;
+    cursor: pointer;
+    transition: background-color 0.3s;
+}
+
+button:hover {
+    background-color: #0056b3;
+}

templates/index.html

@@ -0,0 +1,25 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+    <meta charset="UTF-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1.0">
+    <title>Textual Similarity Predictor</title>
+    <link rel="stylesheet" href="{{ url_for('static', filename='styles.css') }}">
+</head>
+<body>
+    <div class="container">
+        <h1>Textual Similarity Predictor</h1>
+        <form id="predictForm" action="/predict" method="post">
+            <div class="input-group">
+                <label for="sentence1">Enter Sentence 1:</label>
+                <input type="text" id="sentence1" name="sentence1" required>
+            </div>
+            <div class="input-group">
+                <label for="sentence2">Enter Sentence 2:</label>
+                <input type="text" id="sentence2" name="sentence2" required>
+            </div>
+            <button type="submit">Predict Similarity</button>
+        </form>
+    </div>
+</body>
+</html>

templates/result.html

@@ -0,0 +1,18 @@
+
+<!DOCTYPE html>
+<html lang="en">
+<head>
+    <meta charset="UTF-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1.0">
+    <title>Similarity Result</title>
+    <link rel="stylesheet" href="{{ url_for('static', filename='styles.css') }}">
+</head>
+<body>
+    <div class="container">
+        <h1>Textual Similarity Result</h1>
+        <p><strong>Sentence 1:</strong> {{ sentence1 }}</p>
+        <p><strong>Sentence 2:</strong> {{ sentence2 }}</p>
+        <p><strong>Similarity Score:</strong> {{ similarity_score }}</p>
+    </div>
+</body>
+</html>