Spaces:
Running
Running
Commit
·
d85e33e
1
Parent(s):
4667b7d
feat: add max tokens control for model responses
Browse files
app.py
CHANGED
@@ -24,14 +24,14 @@ AVAILABLE_MODELS = [
|
|
24 |
# Initialize inference client
|
25 |
inference_client = InferenceClient(token=HF_TOKEN)
|
26 |
|
27 |
-
async def get_model_response(prompt, model_name, temperature_value, do_sample):
|
28 |
"""Get response from a Hugging Face model."""
|
29 |
try:
|
30 |
# Build kwargs dynamically
|
31 |
generation_args = {
|
32 |
"prompt": prompt,
|
33 |
"model": model_name,
|
34 |
-
"max_new_tokens":
|
35 |
"do_sample": do_sample,
|
36 |
"return_full_text": False
|
37 |
}
|
@@ -46,18 +46,23 @@ async def get_model_response(prompt, model_name, temperature_value, do_sample):
|
|
46 |
None,
|
47 |
partial(inference_client.text_generation, **generation_args)
|
48 |
)
|
|
|
|
|
|
|
|
|
|
|
49 |
return response
|
50 |
|
51 |
except Exception as e:
|
52 |
return f"Error: {str(e)}"
|
53 |
|
54 |
-
async def process_single_response(prompt, model_name, temp, do_sample, chatbot):
|
55 |
"""Process a single model response and update its chatbot."""
|
56 |
-
response = await get_model_response(prompt, model_name, temp, do_sample)
|
57 |
chat_history = [{"role": "user", "content": prompt}, {"role": "assistant", "content": response}]
|
58 |
return chat_history
|
59 |
|
60 |
-
async def compare_models(prompt, model1, model2, temp1, temp2, do_sample1, do_sample2):
|
61 |
"""Compare outputs from two selected models."""
|
62 |
if not prompt.strip():
|
63 |
empty_response = [{"role": "user", "content": prompt}, {"role": "assistant", "content": "Please enter a prompt"}]
|
@@ -69,8 +74,8 @@ async def compare_models(prompt, model1, model2, temp1, temp2, do_sample1, do_sa
|
|
69 |
yield initial_message, initial_message, gr.update(interactive=False)
|
70 |
|
71 |
# Create tasks for both model responses
|
72 |
-
task1 = asyncio.create_task(process_single_response(prompt, model1, temp1, do_sample1, "chatbot1"))
|
73 |
-
task2 = asyncio.create_task(process_single_response(prompt, model2, temp2, do_sample2, "chatbot2"))
|
74 |
|
75 |
chat1 = chat2 = initial_message
|
76 |
start_time = asyncio.get_event_loop().time()
|
@@ -162,6 +167,13 @@ with gr.Blocks(css="""
|
|
162 |
interactive=False,
|
163 |
elem_classes=["disabled-slider"]
|
164 |
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
165 |
chatbot1 = gr.Chatbot(
|
166 |
label="Model 1 Output",
|
167 |
show_label=True,
|
@@ -188,6 +200,13 @@ with gr.Blocks(css="""
|
|
188 |
interactive=False,
|
189 |
elem_classes=["disabled-slider"]
|
190 |
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
191 |
chatbot2 = gr.Chatbot(
|
192 |
label="Model 2 Output",
|
193 |
show_label=True,
|
@@ -206,7 +225,7 @@ with gr.Blocks(css="""
|
|
206 |
queue=False
|
207 |
).then(
|
208 |
fn=compare_models,
|
209 |
-
inputs=[prompt, model1_dropdown, model2_dropdown, temp1, temp2, do_sample1, do_sample2],
|
210 |
outputs=[chatbot1, chatbot2, submit_btn],
|
211 |
queue=True # Enable queuing for streaming updates
|
212 |
)
|
|
|
24 |
# Initialize inference client
|
25 |
inference_client = InferenceClient(token=HF_TOKEN)
|
26 |
|
27 |
+
async def get_model_response(prompt, model_name, temperature_value, do_sample, max_tokens):
|
28 |
"""Get response from a Hugging Face model."""
|
29 |
try:
|
30 |
# Build kwargs dynamically
|
31 |
generation_args = {
|
32 |
"prompt": prompt,
|
33 |
"model": model_name,
|
34 |
+
"max_new_tokens": max_tokens,
|
35 |
"do_sample": do_sample,
|
36 |
"return_full_text": False
|
37 |
}
|
|
|
46 |
None,
|
47 |
partial(inference_client.text_generation, **generation_args)
|
48 |
)
|
49 |
+
|
50 |
+
# Check if response might be truncated
|
51 |
+
if len(response) >= max_tokens * 4: # Rough estimate of tokens to characters ratio
|
52 |
+
response += "\n\n[Warning: Response may have been truncated. Try increasing the max tokens if the response seems incomplete.]"
|
53 |
+
|
54 |
return response
|
55 |
|
56 |
except Exception as e:
|
57 |
return f"Error: {str(e)}"
|
58 |
|
59 |
+
async def process_single_response(prompt, model_name, temp, do_sample, max_tokens, chatbot):
|
60 |
"""Process a single model response and update its chatbot."""
|
61 |
+
response = await get_model_response(prompt, model_name, temp, do_sample, max_tokens)
|
62 |
chat_history = [{"role": "user", "content": prompt}, {"role": "assistant", "content": response}]
|
63 |
return chat_history
|
64 |
|
65 |
+
async def compare_models(prompt, model1, model2, temp1, temp2, do_sample1, do_sample2, max_tokens1, max_tokens2):
|
66 |
"""Compare outputs from two selected models."""
|
67 |
if not prompt.strip():
|
68 |
empty_response = [{"role": "user", "content": prompt}, {"role": "assistant", "content": "Please enter a prompt"}]
|
|
|
74 |
yield initial_message, initial_message, gr.update(interactive=False)
|
75 |
|
76 |
# Create tasks for both model responses
|
77 |
+
task1 = asyncio.create_task(process_single_response(prompt, model1, temp1, do_sample1, max_tokens1, "chatbot1"))
|
78 |
+
task2 = asyncio.create_task(process_single_response(prompt, model2, temp2, do_sample2, max_tokens2, "chatbot2"))
|
79 |
|
80 |
chat1 = chat2 = initial_message
|
81 |
start_time = asyncio.get_event_loop().time()
|
|
|
167 |
interactive=False,
|
168 |
elem_classes=["disabled-slider"]
|
169 |
)
|
170 |
+
max_tokens1 = gr.Slider(
|
171 |
+
label="Maximum new tokens in response",
|
172 |
+
minimum=10,
|
173 |
+
maximum=2000,
|
174 |
+
step=10,
|
175 |
+
value=100
|
176 |
+
)
|
177 |
chatbot1 = gr.Chatbot(
|
178 |
label="Model 1 Output",
|
179 |
show_label=True,
|
|
|
200 |
interactive=False,
|
201 |
elem_classes=["disabled-slider"]
|
202 |
)
|
203 |
+
max_tokens2 = gr.Slider(
|
204 |
+
label="Maximum new tokens in response",
|
205 |
+
minimum=10,
|
206 |
+
maximum=2000,
|
207 |
+
step=10,
|
208 |
+
value=100
|
209 |
+
)
|
210 |
chatbot2 = gr.Chatbot(
|
211 |
label="Model 2 Output",
|
212 |
show_label=True,
|
|
|
225 |
queue=False
|
226 |
).then(
|
227 |
fn=compare_models,
|
228 |
+
inputs=[prompt, model1_dropdown, model2_dropdown, temp1, temp2, do_sample1, do_sample2, max_tokens1, max_tokens2],
|
229 |
outputs=[chatbot1, chatbot2, submit_btn],
|
230 |
queue=True # Enable queuing for streaming updates
|
231 |
)
|