Update app.py

app.py

@@ -6,7 +6,13 @@ import sys
 import json
 import base64
 import tqdm
+import base64
+
 
+# Function to encode the image
+def encode_image(image_path):
+    with open(image_path, "rb") as image_file:
+        return base64.b64encode(image_file.read()).decode("utf-8")
 
 from openai import OpenAI
 # 设置 OpenAI 的 API 密钥和 API 基础 URL 以使用 vLLM 的 API 服务器。
@@ -20,118 +26,60 @@ client = OpenAI(
 )
 
 
-def convert_sigle_message_to_gpt(dialogs):
-    if not dialogs:
-        return []
-
-    merged = []
-    current_role = None
-    current_contents = []
-
-    for item in dialogs:
-        role = item["role"]
-        content = item["content"]
-
-        # 如果 role 变化了，说明要开启一个新的合并段
-        if role != current_role:
-            # 如果之前有积累的内容，需要写入 merged
-            if current_role is not None:
-                merged.append({
-                    "role": current_role,
-                    "content": current_contents
-                })
-            # 重置当前合并信息
-            current_role = role
-            current_contents = []
-            if isinstance(content, tuple):
-                for path in content:
-                    current_contents.append({
-                            "type": "image_url",
-                            "image_url": {"url": f"data:image/jpeg;base64,{encode_image(path)}"}})
-            else:
-                current_contents.append({"type": "text", "text": content})
-        else:
-            # 如果 role 相同，则把 content 加到当前内容列表里
-            if isinstance(content, tuple):
-                for path in content:
-                    current_contents.append({
-                            "type": "image_url",
-                            "image_url": {"url": f"data:image/jpeg;base64,{encode_image(path)}"}})
-            else:
-                current_contents.append({"type": "text", "text": content})
-
-    # 循环结束后，把最后的合并段写入 merged
-    if current_role is not None:
-        merged.append({
-            "role": current_role,
-            "content": current_contents
-        })
-
-    return merged
-
-def clear_fn():
-    return []
-
-def store_values(num_val, name_val):
-    return num_val, name_val  # 存入状态
-
-def use_values(num_state, name_state):
-    return f"使用了数量: {num_state}，文件名: {name_state}"
 
-def main():
-    
- 
-    
-    def bot(history: list):
-        history = convert_sigle_message_to_gpt(history)
-        response = openai.chat.completions.create(
-            model=self.model_name,
-            messages=history,
-            extra_body={},
-            extra_headers={
-                "apikey": "empty"
-            },
-            stream=True,
-            temperature=0.7,
-            top_p=1.0,
-        )
-
-        history.append({"role": "assistant", "content": ""})
-        for character in response:
-            if not character:
-                continue
-            history[-1]["content"] += character
-            yield history
-
-    # 页面布局
+def run_example(image, text_input=None, model_id=None):
+    messages=[
+        {"role": "user", 
+         "content": 
+            [
+            "type": "image_url","image_url": {"url": f"data:image/jpeg;base64,{encode_image(image)}"},
+            {"type": "text", "text": text_input} if text_input is not Noen,
+                
+            ]
+        }
+    ]
     
-    with gr.Blocks() as demo:
+    response = openai.chat.completions.create(
+                model=self.model_name,
+                messages=messages,
+                extra_body=self.extra,
+                extra_headers={
+                    "apikey": self.apikey
+                },
+                stream=self.stream,
+                temperature=0.7,
+                top_p=1.0,
+            )
+    for chunk in response:
+        yield chunk.choices[0].delta.content
+
+
+
+css = """
+  #output {
+    height: 500px; 
+    overflow: auto; 
+    border: 1px solid #ccc; 
+  }
+"""
+
+with gr.Blocks(css=css) as demo:
+    gr.Markdown(DESCRIPTION)
+    with gr.Tab(label="MiMo-VL-7B Input"):
         with gr.Row():
-            with gr.Column(scale=4):
-                chatbot = gr.Chatbot(elem_id="chatbot", bubble_full_width=True, type="messages")
-                chat_input = gr.MultimodalTextbox(
-                    interactive=True,
-                    file_count="multiple",
-                    placeholder="Enter message or upload file...",
-                    show_label=False,
-                    sources=["microphone", "upload"],
-                    elem_id="chat_page"
-                )
-                clear_button = gr.Button('Clear')
-                chat_msg = chat_input.submit(
-                    add_message, [chatbot, chat_input], [chatbot, chat_input]
-                )
-                
-                bot_msg = chat_msg.then(bot, chatbot, chatbot, api_name="bot_response")
-                bot_msg.then(lambda: gr.MultimodalTextbox(interactive=True), None, [chat_input])
-
-                chatbot.like(print_like_dislike, None, None, like_user_message=True)       
-
-        clear_button.click(fn=clear_fn, inputs=[], outputs=chatbot)
+            with gr.Column():
+                input_img = gr.Image(label="Input Picture")
+                model_selector = gr.Dropdown(choices=list(models.keys()), 
+                                             label="Model", 
+                                             value="XiaomiMiMo/MiMo-VL-7B-SFT")
+                text_input = gr.Textbox(label="Text Prompt")
+                submit_btn = gr.Button(value="Submit")
+            with gr.Column():
+                output_text = gr.Textbox(label="Output Text")
         
-    demo.launch(share=True)
+        submit_btn.click(run_example, [input_img, text_input, model_selector], [output_text])
+
+demo.queue(api_open=False)
+demo.launch(debug=True)
 
-    
-if __name__ == "__main__":
-    main()