Update app.py

app.py

@@ -1,36 +1,50 @@
 import gradio as gr
 from transformers import pipeline, AutoTokenizer, AutoModelForCausalLM
 
-# Charger le modèle et le tokenizer
+# Load the model and tokenizer
 tokenizer = AutoTokenizer.from_pretrained("unsloth/Llama-3.2-1B")
 model = AutoModelForCausalLM.from_pretrained("unsloth/Llama-3.2-1B")
 
-# Utiliser une pipeline pour la génération de texte
+# Use a pipeline for text generation
 text_gen_pipeline = pipeline("text-generation", model=model, tokenizer=tokenizer)
 
-# Fonction de génération de texte
-def generate_text(prompt, max_length=50):
-    generated_text = text_gen_pipeline(prompt, max_length=max_length, num_return_sequences=1)
+# Text generation function with additional parameters
+def generate_text(prompt, max_length=50, temperature=0.7, top_p=0.9, top_k=50):
+    generated_text = text_gen_pipeline(prompt, 
+                                       max_length=max_length, 
+                                       temperature=temperature, 
+                                       top_p=top_p, 
+                                       top_k=top_k, 
+                                       num_return_sequences=1)
     return generated_text[0]['generated_text']
 
-# Interface Gradio
+# Gradio Interface
 with gr.Blocks() as demo:
-    gr.Markdown("## Génération de texte avec Llama 3.2 - 1B")
+    gr.Markdown("## Text Generation with Llama 3.2 - 1B")
     
-    # Entrée du texte par l'utilisateur
-    prompt_input = gr.Textbox(label="Entrée (prompt)", placeholder="Entrez votre texte ici...")
+    # Input box for user prompt
+    prompt_input = gr.Textbox(label="Input (Prompt)", placeholder="Enter your prompt here...")
     
-    # Slider pour la longueur maximale de la génération
-    max_length_input = gr.Slider(minimum=10, maximum=200, value=50, step=10, label="Longueur maximale")
+    # Slider for maximum text length
+    max_length_input = gr.Slider(minimum=10, maximum=200, value=50, step=10, label="Maximum Length")
     
-    # Zone de sortie pour le texte généré
-    output_text = gr.Textbox(label="Texte généré")
+    # Slider for temperature (controls creativity)
+    temperature_input = gr.Slider(minimum=0.1, maximum=1.0, value=0.7, step=0.1, label="Temperature (creativity)")
     
-    # Bouton de soumission
-    generate_button = gr.Button("Générer")
+    # Slider for top_p (nucleus sampling)
+    top_p_input = gr.Slider(minimum=0.1, maximum=1.0, value=0.9, step=0.1, label="Top-p (nucleus sampling)")
     
-    # Action sur le bouton
-    generate_button.click(generate_text, inputs=[prompt_input, max_length_input], outputs=output_text)
+    # Slider for top_k (controls diversity)
+    top_k_input = gr.Slider(minimum=1, maximum=100, value=50, step=1, label="Top-k (sampling diversity)")
+    
+    # Output box for the generated text
+    output_text = gr.Textbox(label="Generated Text")
+    
+    # Submit button
+    generate_button = gr.Button("Generate")
+    
+    # Action on button click
+    generate_button.click(generate_text, inputs=[prompt_input, max_length_input, temperature_input, top_p_input, top_k_input], outputs=output_text)
 
-# Lancer l'application
+# Launch the app
 demo.launch()