application deployed

app.py

@@ -9,19 +9,25 @@
 #######################
 
 import gradio as gr
-from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
+from transformers import AutoTokenizer, AutoModelForCausalLM
+import torch
 
+# Model ID (7B is safer for free-tier Space)
 model_id = "codellama/CodeLlama-7b-Instruct-hf"
 
-# Load the tokenizer and model
+# Load tokenizer and model
 tokenizer = AutoTokenizer.from_pretrained(model_id)
-model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto", torch_dtype="auto")
+model = AutoModelForCausalLM.from_pretrained(
+    model_id, 
+    device_map="auto", 
+    torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32
+)
 
-# Use text-generation pipeline
-generator = pipeline("text-generation", model=model, tokenizer=tokenizer)
+# Ensure pad token is set to eos token to avoid padding warnings
+tokenizer.pad_token = tokenizer.eos_token
 
-# Define the function to handle the conversion
 def convert_python_to_r(python_code):
+    # Prompt construction
     prompt = f"""### Task:
 Convert the following Python code to equivalent R code.
 
@@ -29,22 +35,40 @@ Convert the following Python code to equivalent R code.
 {python_code}
 
 ### R code:"""
-    #result = generator(prompt, max_length=1024, temperature=0.2, num_return_sequences=1)
-    result = generator(
-    prompt, 
-     max_length=1024, 
-     temperature=0.2, 
-     num_return_sequences=1, 
-     do_sample=True,                          # <-- Add this!
-     pad_token_id=tokenizer.eos_token_id      # <-- Safely set pad_token_id
+
+    # Tokenize input with truncation
+    input_ids = tokenizer(prompt, return_tensors="pt", truncation=True).input_ids
+
+    # Move input to GPU if available
+    if torch.cuda.is_available():
+        input_ids = input_ids.to("cuda")
+
+    # Generate output
+    outputs = model.generate(
+        input_ids,
+        max_length=1024,
+        do_sample=True,
+        temperature=0.2,
+        pad_token_id=tokenizer.eos_token_id,
+        num_return_sequences=1
     )
-    return result[0]['generated_text']
+
+    # Decode the generated output
+    generated_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
+
+    # Remove the prompt part from output, keep only R code
+    # Optional: Try to split from "### R code:"
+    if "### R code:" in generated_text:
+        generated_text = generated_text.split("### R code:")[-1].strip()
+
+    return generated_text
 
 # Gradio Interface
 gr.Interface(
     fn=convert_python_to_r,
     inputs=gr.Textbox(lines=10, placeholder="Paste your Python code here..."),
     outputs="text",
-    title="Python to R Code Converter using CodeLlama",
-    description="Enter Python code, and this tool will convert it into equivalent R code using CodeLlama Instruct model."
+    title="Python to R Code Converter using CodeLlama 7B Instruct",
+    description="Enter Python code below, and the tool will convert it to R code using the CodeLlama 7B Instruct model."
 ).launch()
+