Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
@@ -93,16 +93,14 @@ def push_feedback_to_hub(hf_token=None):
|
|
93 |
print(f"Error pushing feedback data to Hub: {e}")
|
94 |
return False
|
95 |
|
96 |
-
#
|
97 |
-
chat_history_state = []
|
98 |
-
|
99 |
@spaces.GPU(duration=120)
|
100 |
-
def predict(message, history, temperature, top_p):
|
101 |
-
|
102 |
-
|
103 |
-
# Update our chat history state
|
104 |
history.append({"role": "user", "content": message})
|
105 |
-
|
|
|
|
|
106 |
|
107 |
input_text = tokenizer.apply_chat_template(history, tokenize=False, add_generation_prompt=True)
|
108 |
inputs = tokenizer.encode(input_text, return_tensors="pt").to(device)
|
@@ -129,18 +127,18 @@ def predict(message, history, temperature, top_p):
|
|
129 |
partial_text = ""
|
130 |
for new_text in streamer:
|
131 |
partial_text += new_text
|
132 |
-
yield partial_text
|
133 |
|
134 |
-
# After generation
|
135 |
-
|
|
|
|
|
136 |
|
137 |
# Function to handle the research feedback submission
|
138 |
-
def submit_research_feedback(satisfaction, feedback_text):
|
139 |
"""Save user feedback both locally and to HuggingFace Hub"""
|
140 |
-
global chat_history_state
|
141 |
-
|
142 |
# Save locally first
|
143 |
-
feedback_id = save_feedback_locally(
|
144 |
|
145 |
# Get token from environment variable
|
146 |
env_token = os.environ.get("HF_TOKEN")
|
@@ -155,16 +153,29 @@ def submit_research_feedback(satisfaction, feedback_text):
|
|
155 |
|
156 |
return status_msg
|
157 |
|
158 |
-
# Create the Gradio interface
|
159 |
with gr.Blocks() as demo:
|
|
|
|
|
|
|
160 |
with gr.Row():
|
161 |
with gr.Column(scale=3):
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
162 |
chatbot = gr.ChatInterface(
|
163 |
-
|
164 |
additional_inputs=[
|
|
|
165 |
gr.Slider(0.1, 2.0, value=0.7, step=0.1, label="Temperature"),
|
166 |
gr.Slider(0.1, 1.0, value=0.9, step=0.05, label="Top-P")
|
167 |
],
|
|
|
168 |
type="messages"
|
169 |
)
|
170 |
|
@@ -199,10 +210,10 @@ with gr.Blocks() as demo:
|
|
199 |
feedback_modal
|
200 |
)
|
201 |
|
202 |
-
# Connect the submit button to the submit_research_feedback function with the current
|
203 |
submit_button.click(
|
204 |
-
|
205 |
-
inputs=[satisfaction, feedback_text],
|
206 |
outputs=response_text
|
207 |
)
|
208 |
|
|
|
93 |
print(f"Error pushing feedback data to Hub: {e}")
|
94 |
return False
|
95 |
|
96 |
+
# Modified predict function to update conversation state
|
|
|
|
|
97 |
@spaces.GPU(duration=120)
|
98 |
+
def predict(message, history, state, temperature, top_p):
|
99 |
+
# Update history with user message
|
|
|
|
|
100 |
history.append({"role": "user", "content": message})
|
101 |
+
|
102 |
+
# Update the conversation state
|
103 |
+
state = history.copy()
|
104 |
|
105 |
input_text = tokenizer.apply_chat_template(history, tokenize=False, add_generation_prompt=True)
|
106 |
inputs = tokenizer.encode(input_text, return_tensors="pt").to(device)
|
|
|
127 |
partial_text = ""
|
128 |
for new_text in streamer:
|
129 |
partial_text += new_text
|
130 |
+
yield partial_text, state
|
131 |
|
132 |
+
# After full generation, update state with assistant's response
|
133 |
+
history.append({"role": "assistant", "content": partial_text})
|
134 |
+
state = history.copy()
|
135 |
+
return partial_text, state
|
136 |
|
137 |
# Function to handle the research feedback submission
|
138 |
+
def submit_research_feedback(conversation_state, satisfaction, feedback_text):
|
139 |
"""Save user feedback both locally and to HuggingFace Hub"""
|
|
|
|
|
140 |
# Save locally first
|
141 |
+
feedback_id = save_feedback_locally(conversation_state, satisfaction, feedback_text)
|
142 |
|
143 |
# Get token from environment variable
|
144 |
env_token = os.environ.get("HF_TOKEN")
|
|
|
153 |
|
154 |
return status_msg
|
155 |
|
156 |
+
# Create the Gradio blocks interface
|
157 |
with gr.Blocks() as demo:
|
158 |
+
# State to track conversation history
|
159 |
+
conversation_state = gr.State([])
|
160 |
+
|
161 |
with gr.Row():
|
162 |
with gr.Column(scale=3):
|
163 |
+
# Custom chat function wrapper to update state
|
164 |
+
def chat_with_state(message, history, state, temperature, top_p):
|
165 |
+
for partial_response, updated_state in predict(message, history, state, temperature, top_p):
|
166 |
+
# Update our state with each yield
|
167 |
+
state = updated_state
|
168 |
+
yield partial_response, state
|
169 |
+
|
170 |
+
# Create ChatInterface
|
171 |
chatbot = gr.ChatInterface(
|
172 |
+
chat_with_state,
|
173 |
additional_inputs=[
|
174 |
+
conversation_state,
|
175 |
gr.Slider(0.1, 2.0, value=0.7, step=0.1, label="Temperature"),
|
176 |
gr.Slider(0.1, 1.0, value=0.9, step=0.05, label="Top-P")
|
177 |
],
|
178 |
+
additional_outputs=[conversation_state],
|
179 |
type="messages"
|
180 |
)
|
181 |
|
|
|
210 |
feedback_modal
|
211 |
)
|
212 |
|
213 |
+
# Connect the submit button to the submit_research_feedback function with the current conversation state
|
214 |
submit_button.click(
|
215 |
+
submit_research_feedback,
|
216 |
+
inputs=[conversation_state, satisfaction, feedback_text],
|
217 |
outputs=response_text
|
218 |
)
|
219 |
|