Update app.py
Browse files
app.py
CHANGED
@@ -147,13 +147,13 @@
|
|
147 |
|
148 |
import os
|
149 |
import gradio as gr
|
150 |
-
from transformers import
|
151 |
import torch
|
152 |
from typing import List, Dict
|
153 |
import logging
|
154 |
import traceback
|
155 |
|
156 |
-
# Configure detailed logging
|
157 |
logging.basicConfig(
|
158 |
level=logging.INFO,
|
159 |
format='%(asctime)s - %(levelname)s - %(message)s'
|
@@ -162,54 +162,40 @@ logger = logging.getLogger(__name__)
|
|
162 |
|
163 |
class MedicalAssistant:
|
164 |
def __init__(self):
|
165 |
-
"""
|
|
|
|
|
|
|
166 |
try:
|
167 |
logger.info("Starting model initialization...")
|
168 |
|
169 |
-
#
|
170 |
-
self.model_name = "
|
171 |
self.max_length = 2048
|
172 |
self.device = "cuda" if torch.cuda.is_available() else "cpu"
|
173 |
|
|
|
174 |
logger.info(f"Using device: {self.device}")
|
175 |
logger.info(f"Available CUDA devices: {torch.cuda.device_count() if torch.cuda.is_available() else 'None'}")
|
176 |
if torch.cuda.is_available():
|
177 |
logger.info(f"CUDA memory: {torch.cuda.get_device_properties(0).total_memory / 1e9:.2f} GB")
|
178 |
|
179 |
-
#
|
180 |
-
logger.info(
|
181 |
-
|
182 |
-
|
183 |
-
|
184 |
-
|
185 |
-
|
186 |
-
|
187 |
-
|
188 |
-
logger.error(f"Failed to load tokenizer: {str(e)}")
|
189 |
-
logger.error(traceback.format_exc())
|
190 |
-
raise
|
191 |
|
192 |
-
#
|
|
|
|
|
193 |
if self.tokenizer.pad_token is None:
|
194 |
self.tokenizer.pad_token = self.tokenizer.eos_token
|
195 |
-
|
196 |
-
|
197 |
-
# Load model with more conservative settings
|
198 |
-
logger.info("Loading model - this may take a few minutes...")
|
199 |
-
try:
|
200 |
-
self.model = AutoModelForCausalLM.from_pretrained(
|
201 |
-
self.model_name,
|
202 |
-
torch_dtype=torch.float16,
|
203 |
-
device_map="auto",
|
204 |
-
load_in_4bit=True, # More conservative than 8-bit
|
205 |
-
trust_remote_code=True,
|
206 |
-
low_cpu_mem_usage=True
|
207 |
-
)
|
208 |
-
logger.info("Model loaded successfully!")
|
209 |
-
except Exception as e:
|
210 |
-
logger.error(f"Failed to load model: {str(e)}")
|
211 |
-
logger.error(traceback.format_exc())
|
212 |
-
raise
|
213 |
|
214 |
except Exception as e:
|
215 |
logger.error(f"Initialization failed: {str(e)}")
|
@@ -217,43 +203,42 @@ class MedicalAssistant:
|
|
217 |
raise
|
218 |
|
219 |
def generate_response(self, message: str, chat_history: List[Dict] = None) -> str:
|
220 |
-
"""
|
|
|
|
|
|
|
221 |
try:
|
222 |
-
logger.info("
|
223 |
|
224 |
-
# Prepare the
|
225 |
system_prompt = """You are a medical AI assistant. Respond to medical queries
|
226 |
professionally and accurately. If you're unsure, always recommend consulting
|
227 |
with a healthcare provider."""
|
228 |
|
229 |
-
|
230 |
-
|
231 |
-
|
232 |
-
|
233 |
-
|
234 |
-
return_tensors="pt",
|
235 |
-
padding=True,
|
236 |
-
truncation=True,
|
237 |
-
max_length=self.max_length
|
238 |
-
)
|
239 |
|
240 |
-
#
|
241 |
-
|
|
|
242 |
|
243 |
logger.info("Generating response")
|
244 |
-
|
245 |
-
|
246 |
-
|
247 |
-
|
248 |
-
|
249 |
-
|
250 |
-
|
251 |
-
|
252 |
-
|
253 |
-
|
254 |
|
255 |
-
response
|
256 |
-
response = response.split("
|
257 |
|
258 |
logger.info("Response generated successfully")
|
259 |
return response
|
@@ -263,11 +248,14 @@ class MedicalAssistant:
|
|
263 |
logger.error(traceback.format_exc())
|
264 |
return f"I apologize, but I encountered an error: {str(e)}"
|
265 |
|
266 |
-
#
|
267 |
assistant = None
|
268 |
|
269 |
def initialize_assistant():
|
270 |
-
"""
|
|
|
|
|
|
|
271 |
global assistant
|
272 |
try:
|
273 |
logger.info("Attempting to initialize assistant")
|
@@ -280,7 +268,9 @@ def initialize_assistant():
|
|
280 |
return False
|
281 |
|
282 |
def chat_response(message: str, history: List[Dict]):
|
283 |
-
"""
|
|
|
|
|
284 |
global assistant
|
285 |
|
286 |
if assistant is None:
|
@@ -295,11 +285,12 @@ def chat_response(message: str, history: List[Dict]):
|
|
295 |
logger.error(traceback.format_exc())
|
296 |
return f"I encountered an error: {str(e)}"
|
297 |
|
298 |
-
# Create Gradio interface
|
299 |
demo = gr.ChatInterface(
|
300 |
fn=chat_response,
|
301 |
-
title="Medical Assistant (
|
302 |
-
description="This
|
|
|
303 |
examples=[
|
304 |
"What are the symptoms of malaria?",
|
305 |
"How can I prevent type 2 diabetes?",
|
@@ -307,7 +298,7 @@ demo = gr.ChatInterface(
|
|
307 |
]
|
308 |
)
|
309 |
|
310 |
-
# Launch the
|
311 |
if __name__ == "__main__":
|
312 |
logger.info("Starting the application")
|
313 |
demo.launch()
|
|
|
147 |
|
148 |
import os
|
149 |
import gradio as gr
|
150 |
+
from transformers import pipeline, AutoTokenizer, AutoModelForCausalLM
|
151 |
import torch
|
152 |
from typing import List, Dict
|
153 |
import logging
|
154 |
import traceback
|
155 |
|
156 |
+
# Configure detailed logging to help us track the model's behavior
|
157 |
logging.basicConfig(
|
158 |
level=logging.INFO,
|
159 |
format='%(asctime)s - %(levelname)s - %(message)s'
|
|
|
162 |
|
163 |
class MedicalAssistant:
|
164 |
def __init__(self):
|
165 |
+
"""
|
166 |
+
Initialize the medical assistant using a pre-quantized 4-bit model.
|
167 |
+
This approach uses less memory while maintaining good performance.
|
168 |
+
"""
|
169 |
try:
|
170 |
logger.info("Starting model initialization...")
|
171 |
|
172 |
+
# Define model configuration
|
173 |
+
self.model_name = "emircanerol/Llama3-Med42-8B-4bit"
|
174 |
self.max_length = 2048
|
175 |
self.device = "cuda" if torch.cuda.is_available() else "cpu"
|
176 |
|
177 |
+
# Log system information for debugging
|
178 |
logger.info(f"Using device: {self.device}")
|
179 |
logger.info(f"Available CUDA devices: {torch.cuda.device_count() if torch.cuda.is_available() else 'None'}")
|
180 |
if torch.cuda.is_available():
|
181 |
logger.info(f"CUDA memory: {torch.cuda.get_device_properties(0).total_memory / 1e9:.2f} GB")
|
182 |
|
183 |
+
# Initialize the pipeline for text generation
|
184 |
+
logger.info("Initializing text generation pipeline...")
|
185 |
+
self.pipe = pipeline(
|
186 |
+
"text-generation",
|
187 |
+
model=self.model_name,
|
188 |
+
device_map="auto",
|
189 |
+
torch_dtype=torch.float16
|
190 |
+
)
|
191 |
+
logger.info("Pipeline initialized successfully!")
|
|
|
|
|
|
|
192 |
|
193 |
+
# Load tokenizer separately for more control over text processing
|
194 |
+
logger.info("Loading tokenizer...")
|
195 |
+
self.tokenizer = AutoTokenizer.from_pretrained(self.model_name)
|
196 |
if self.tokenizer.pad_token is None:
|
197 |
self.tokenizer.pad_token = self.tokenizer.eos_token
|
198 |
+
logger.info("Tokenizer loaded successfully!")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
199 |
|
200 |
except Exception as e:
|
201 |
logger.error(f"Initialization failed: {str(e)}")
|
|
|
203 |
raise
|
204 |
|
205 |
def generate_response(self, message: str, chat_history: List[Dict] = None) -> str:
|
206 |
+
"""
|
207 |
+
Generate a response using the text generation pipeline.
|
208 |
+
The pipeline handles most of the complexity of text generation for us.
|
209 |
+
"""
|
210 |
try:
|
211 |
+
logger.info("Preparing message for generation")
|
212 |
|
213 |
+
# Prepare the conversation format
|
214 |
system_prompt = """You are a medical AI assistant. Respond to medical queries
|
215 |
professionally and accurately. If you're unsure, always recommend consulting
|
216 |
with a healthcare provider."""
|
217 |
|
218 |
+
# Format messages for the model
|
219 |
+
messages = [
|
220 |
+
{"role": "system", "content": system_prompt},
|
221 |
+
{"role": "user", "content": message}
|
222 |
+
]
|
|
|
|
|
|
|
|
|
|
|
223 |
|
224 |
+
# Convert messages to a format the model expects
|
225 |
+
prompt = "\n".join([f"{msg['role']}: {msg['content']}" for msg in messages])
|
226 |
+
prompt += "\nassistant:"
|
227 |
|
228 |
logger.info("Generating response")
|
229 |
+
# Generate response using the pipeline
|
230 |
+
response = self.pipe(
|
231 |
+
prompt,
|
232 |
+
max_new_tokens=512,
|
233 |
+
do_sample=True,
|
234 |
+
temperature=0.7,
|
235 |
+
top_p=0.95,
|
236 |
+
repetition_penalty=1.1,
|
237 |
+
pad_token_id=self.tokenizer.pad_token_id
|
238 |
+
)[0]["generated_text"]
|
239 |
|
240 |
+
# Extract the assistant's response from the full generated text
|
241 |
+
response = response.split("assistant:")[-1].strip()
|
242 |
|
243 |
logger.info("Response generated successfully")
|
244 |
return response
|
|
|
248 |
logger.error(traceback.format_exc())
|
249 |
return f"I apologize, but I encountered an error: {str(e)}"
|
250 |
|
251 |
+
# Initialize our global assistant
|
252 |
assistant = None
|
253 |
|
254 |
def initialize_assistant():
|
255 |
+
"""
|
256 |
+
Initialize the assistant with error handling and logging.
|
257 |
+
This helps us track any issues during startup.
|
258 |
+
"""
|
259 |
global assistant
|
260 |
try:
|
261 |
logger.info("Attempting to initialize assistant")
|
|
|
268 |
return False
|
269 |
|
270 |
def chat_response(message: str, history: List[Dict]):
|
271 |
+
"""
|
272 |
+
Handle chat messages and maintain conversation context.
|
273 |
+
"""
|
274 |
global assistant
|
275 |
|
276 |
if assistant is None:
|
|
|
285 |
logger.error(traceback.format_exc())
|
286 |
return f"I encountered an error: {str(e)}"
|
287 |
|
288 |
+
# Create the Gradio interface with a clean, professional design
|
289 |
demo = gr.ChatInterface(
|
290 |
fn=chat_response,
|
291 |
+
title="Medical Assistant (4-bit Quantized Version)",
|
292 |
+
description="""This medical assistant uses a 4-bit quantized model for efficient operation.
|
293 |
+
It provides medical guidance while ensuring comprehensive health information gathering.""",
|
294 |
examples=[
|
295 |
"What are the symptoms of malaria?",
|
296 |
"How can I prevent type 2 diabetes?",
|
|
|
298 |
]
|
299 |
)
|
300 |
|
301 |
+
# Launch the application
|
302 |
if __name__ == "__main__":
|
303 |
logger.info("Starting the application")
|
304 |
demo.launch()
|