Update app.py

app.py

@@ -4,24 +4,19 @@ import gradio as gr
 #client = InferenceClient("""K00B404/BagOMistral_14X_Coders-ties-7B""")
 from transformers import AutoTokenizer, AutoModelForSequenceClassification
 import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer
 
-# Replace 'your-username' and 'your-model-name' with your actual username and model name
-tokenizer = AutoTokenizer.from_pretrained('K00B404/Merged_Beowolx-CodePro_Medusa2-7B-Mistral-I-v0-2')
-model = AutoModelForSequenceClassification.from_pretrained('K00B404/Merged_Beowolx-CodePro_Medusa2-7B-Mistral-I-v0-2')
+model_id = 'K00B404/Merged_Beowolx-CodePro_Medusa2-7B-Mistral-I-v0-2'
+tokenizer = AutoTokenizer.from_pretrained(model_id)
 
-# Example input sequence
-input_sequence = "This is an example sentence."
+model = AutoModelForCausalLM.from_pretrained(model_id)
 
-# Tokenize the input sequence
-inputs = tokenizer(input_sequence, return_tensors="pt")
+text = "Hello my name is"
+inputs = tokenizer(text, return_tensors="pt")
 
-# Run the input through the model
-outputs = model(**inputs)
+outputs = model.generate(**inputs, max_new_tokens=20)
+print(tokenizer.decode(outputs[0], skip_special_tokens=True))
 
-# Get the predicted class label
-predicted_class = outputs[0].argmax(-1).item()
-
-print("Predicted class:", predicted_class)
 
 """
 def format_prompt(message, history):