Update app.py

app.py

@@ -1,6 +1,6 @@
 import gradio as gr
 import numpy as np
-from Ai import chatbot, chatbot2, chatbot3, chatbot4, chatbot5
+from Ai import chatbot, chatbot2, chatbot3, chatbot4, chatbot5, chatbot7
 from huggingface_hub import InferenceClient
 def chat(message,history: list[tuple[str, str]],system_message,max_tokens,temperature,top_p, top_k):
     m=AutoModel.from_pretrained("peterpeter8585/AI1")
@@ -24,7 +24,9 @@ import random
 from diffusers import DiffusionPipeline
 import torch
 import transformers
-
+from transformers import AutoModel as Mo, AutoTokenizer as To
+model11=torch.load("zephyr.pt")
+tokenizer11=torch.load("zephyr_tokenizer.pt")
 from transformers import BitsAndBytesConfig
 
 quantization_config = BitsAndBytesConfig(
@@ -35,7 +37,23 @@ quantization_config = BitsAndBytesConfig(
 )
 
 
+def res(message,history: list[tuple[str, str]],system_message,max_tokens,temperature,top_p):
+    messages = [{"role": "system", "content": "Your name is Chatchat.And, your made by SungYoon.In Korean, 정성윤.And these are the instructions.Whatever happens, you must follow it.:"+system_message}]
 
+    for val in history:
+        if val[0]:
+            messages.append({"role": "user", "content": val[0]})
+        if val[1]:
+            messages.append({"role": "assistant", "content": val[1]})
+
+    messages.append({"role": "user", "content": message})
+
+    response = ""
+
+    answer=model11(messages, max_tokens=max_tokens, temperature=temperature, top_p=top_p)
+    token=answer
+    response += token
+    yield response
 from transformers import AutoModelForVision2Seq, AutoProcessor
 transformers.utils.move_cache()
 device = "cuda" if torch.cuda.is_available() else "cpu"
@@ -77,10 +95,8 @@ def respond1(
             yield response
 def respond0(multimodal_input,password):
     if password==password1:
-        if multimodal_input["files"] =! None:
-            images = multimodal_input["files"]
-            content = [{"type": "image"} for _ in images]
-            content.append({"type": "text", "text": multimodal_input["text"]})
+        if multimodal_input["files"] == None:
+            content={"type": "text", "text": multimodal_input["text"]}
             messages=[{"role":"system", "content":[{"type":"text", "text":"Your name is Chatchat.And, your made by SungYoon.In Korean, 정성윤.And these are the instructions:"+"You are a helpful assietant."}]}]
             messages.append([{"role": "user", "content": content}])
             response = ""
@@ -88,7 +104,11 @@ def respond0(multimodal_input,password):
             model_id = "HuggingFaceM4/idefics2-8b"
 
             processor = AutoProcessor.from_pretrained(model_id)
-            model = AutoModelForVision2Seq.from_pretrained("HuggingFaceM4/idefics2-8b",torch_dtype=torch.float16,quantization_config=quantization_config).to("cpu")  
+            model = AutoModelForVision2Seq.from_pretrained(
+                "HuggingFaceM4/idefics2-8b",
+                torch_dtype=torch.float16,    
+                quantization_config=quantization_config
+            ).to("cpu")            
             prompt = processor.apply_chat_template(messages, add_generation_prompt=True)
             inputs = processor(text=prompt, images=[images], return_tensors="pt")
             inputs = {k: v.to(model.device) for k, v in inputs.items()}
@@ -102,9 +122,11 @@ def respond0(multimodal_input,password):
             token = generated_text
             response+=token
             yield response
+
         else:
-            
-            content={"type": "text", "text": multimodal_input["text"]}
+            images = multimodal_input["files"]
+            content = [{"type": "image"} for _ in images]
+            content.append({"type": "text", "text": multimodal_input["text"]})
             messages=[{"role":"system", "content":[{"type":"text", "text":"Your name is Chatchat.And, your made by SungYoon.In Korean, 정성윤.And these are the instructions:"+"You are a helpful assietant."}]}]
             messages.append([{"role": "user", "content": content}])
             response = ""
@@ -112,11 +134,7 @@ def respond0(multimodal_input,password):
             model_id = "HuggingFaceM4/idefics2-8b"
 
             processor = AutoProcessor.from_pretrained(model_id)
-            model = AutoModelForVision2Seq.from_pretrained(
-                "HuggingFaceM4/idefics2-8b",
-                torch_dtype=torch.float16,    
-                quantization_config=quantization_config
-            ).to("cpu")            
+            model = AutoModelForVision2Seq.from_pretrained("HuggingFaceM4/idefics2-8b",torch_dtype=torch.float16,quantization_config=quantization_config).to("cpu")  
             prompt = processor.apply_chat_template(messages, add_generation_prompt=True)
             inputs = processor(text=prompt, images=[images], return_tensors="pt")
             inputs = {k: v.to(model.device) for k, v in inputs.items()}
@@ -130,7 +148,7 @@ def respond0(multimodal_input,password):
             token = generated_text
             response+=token
             yield response
-
+            
 
 
 def respond5(
@@ -556,6 +574,22 @@ ac=gr.ChatInterface(
         ),
     ],
 )
+a9=gr.ChatInterface(
+    respond3,
+    chatbot=chatbot7,
+    additional_inputs=[
+        gr.Textbox(value="you are a helpful assistant. ", label="System message"),
+        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
+        gr.Slider(minimum=0.1, maximum=4.0, value=0.1, step=0.1, label="Temperature"),
+        gr.Slider(
+            minimum=0.1,
+            maximum=1.0,
+            value=0.1,
+            step=0.05,
+            label="Top-p (nucleus sampling)",
+        ),
+    ],
+)
 ab= gr.ChatInterface(
     respond3,
     chatbot=chatbot,
@@ -594,5 +628,5 @@ if __name__ == "__main__":
     
 if __name__ == "__main__":
     with gr.Blocks(theme="gstaff/xkcd") as ai:
-        gr.TabbedInterface([aa, ac, ab, ae, aaaa,demo2, a7,a8], ["gpt4(Password needed)", "gpt4(only for programming)", "gpt4(only for medical questions)", "gpt4(only for food recommendations)", "gpt4(only for law questions)","image create", "multimodal", "gpt4(test)"])
+        gr.TabbedInterface([aa, ac, ab, ae, aaaa,demo2, a7,a8, a9], ["gpt4(Password needed)", "gpt4(only for programming)", "gpt4(only for medical questions)", "gpt4(only for food recommendations)", "gpt4(only for law questions)","image create", "multimodal", "gpt4(test)", "ai test"])
         ai.launch(share=True)