Update app.py

app.py

@@ -1,63 +1,64 @@
-import gradio as gr
 from huggingface_hub import InferenceClient
+import gradio as gr
+import json
+import PyPDF2
 
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
-
-
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+client = InferenceClient(
+    "mistralai/Mistral-7B-Instruct-v0.3"
+)
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+def format_prompt(mode, message, instructions, history):
+    prompt = f"<s>[MODE] {mode} [/MODE] "
+    for user_prompt, bot_response in history:
+        prompt += f"[INST] {user_prompt} [/INST] {bot_response}</s> "
+    prompt += f"[INST] {message} {instructions} [/INST]"
+    return prompt
 
-    messages.append({"role": "user", "content": message})
+def process_input(file, file_type):
+    if file_type == 'pdf':
+        reader = PyPDF2.PdfFileReader(file.name)
+        text = ""
+        for page in range(reader.numPages):
+            text += reader.getPage(page).extractText()
+        return text
+    elif file_type == 'json':
+        data = json.load(file)
+        return json.dumps(data, indent=2)
+    return file
 
-    response = ""
+def generate(mode, file, file_type, instructions, history, temperature=0.9, max_new_tokens=256, top_p=0.95, repetition_penalty=1.0):
+    temperature = max(float(temperature), 1e-2)
+    top_p = float(top_p)
+    
+    text_input = process_input(file, file_type)
+    formatted_prompt = format_prompt(mode, text_input, instructions, history)
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
+    generate_kwargs = dict(
         temperature=temperature,
+        max_new_tokens=max_new_tokens,
         top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
-
-        response += token
-        yield response
-
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", 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=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
-)
+        repetition_penalty=repetition_penalty,
+        do_sample=True,
+        seed=42,
+    )
 
+    stream = client.text_generation(formatted_prompt, **generate_kwargs, stream=True, details=True, return_full_text=False)
+    output = ""
 
-if __name__ == "__main__":
-    demo.launch()
+    for response in stream:
+        output += response.token.text
+        yield output
+    return output
+
+gr.ChatInterface(
+    fn=generate,
+    inputs=[
+        gr.Dropdown(label="Mode", choices=["translation", "summary", "explanation"], value="translation"),
+        gr.File(label="Input File", type="file"),
+        gr.Radio(label="File Type", choices=["pdf", "json"], value="pdf"),
+        gr.Textbox(label="Additional Instructions", placeholder="Enter any additional instructions here"),
+        gr.Chatbot()
+    ],
+    outputs=gr.Chatbot(),
+    title="Mistral 7B v0.3"
+).launch(show_api=False)