Update app.py

app.py

@@ -2,46 +2,46 @@ from transformers import AutoModelForCausalLM, AutoTokenizer
 import gradio as gr
 import torch
 
+
 title = "EZChat"
-description = "A State-of-the-Art Large-scale Pretrained Response generation model Qwen's 7B-Chat"
+description = "A State-of-the-Art Large-scale Pretrained Response generation model (DialoGPT)"
 examples = [["How are you?"]]
 
-tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen-7B-Chat", trust_remote_code=True)
-model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen-7B-Chat", trust_remote_code=True).eval()
-
-history = []  # Initialize chat history
 
-def predict(input, history=history):
-    if input is not None and tokenizer.eos_token is not None:
-        combined_input = input + tokenizer.eos_token
-        new_user_input_ids = tokenizer.encode(combined_input, return_tensors="pt")
+tokenizer = AutoTokenizer.from_pretrained("PygmalionAI/pygmalion-2-7b")
+model = AutoModelForCausalLM.from_pretrained("PygmalionAI/pygmalion-2-7b")
 
-        # Append the new user input tokens to the chat history
-        bot_input_ids = torch.cat([torch.LongTensor(history), new_user_input_ids], dim=-1)
 
-        # Generate a response
-        generated_response_ids = model.generate(
-            bot_input_ids, max_length=20, pad_token_id=tokenizer.eos_token_id
-        )
+def predict(input, history=[]):
+    # tokenize the new input sentence
+    new_user_input_ids = tokenizer.encode(
+        input + tokenizer.eos_token, return_tensors="pt"
+    )
 
-        # Convert the generated response tokens to text
-        response = tokenizer.decode(generated_response_ids[0], skip_special_tokens=True)
+    # append the new user input tokens to the chat history
+    bot_input_ids = torch.cat([torch.LongTensor(history), new_user_input_ids], dim=-1)
 
-        # Append the user input and generated response to the chat history
-        history.extend(new_user_input_ids[0].tolist())
-        history.extend(generated_response_ids[0].tolist())
+    # generate a response
+    history = model.generate(
+        bot_input_ids, max_length=4000, pad_token_id=tokenizer.eos_token_id
+    ).tolist()
 
-        return response, history
+    # convert the tokens to text, and then split the responses into lines
+    response = tokenizer.decode(history[0]).split("<|endoftext|>")
+    # print('decoded_response-->>'+str(response))
+    response = [
+        (response[i], response[i + 1]) for i in range(0, len(response) - 1, 2)
+    ]  # convert to tuples of list
+    # print('response-->>'+str(response))
+    return response, history
 
-    else:
-        print("Input or eos_token is None. Cannot concatenate.")
 
 gr.Interface(
     fn=predict,
     title=title,
     description=description,
     examples=examples,
-    inputs=["text", "text"],
-    outputs=["text", "text"],
+    inputs=["text", "state"],
+    outputs=["chatbot", "state"],
     theme="ParityError/Anime",
-).launch()
+).launch()