Spaces:
Sleeping
Sleeping
Ani07-05
commited on
Commit
Β·
7b9bc7a
1
Parent(s):
92cd1f8
Switch to WiroAI-Finance-Qwen-1.5B model
Browse files
app.py
CHANGED
@@ -1,121 +1,139 @@
|
|
1 |
import streamlit as st
|
2 |
import os
|
3 |
-
from transformers import pipeline
|
4 |
-
import torch
|
5 |
|
6 |
# --- Set Page Config FIRST ---
|
7 |
-
st.set_page_config(layout="wide")
|
8 |
|
9 |
# --- Configuration ---
|
10 |
-
MODEL_NAME = "AdaptLLM/finance-LLM"
|
11 |
-
|
12 |
HF_TOKEN = os.environ.get("HF_TOKEN")
|
13 |
|
14 |
# --- Model Loading (Cached by Streamlit for efficiency) ---
|
15 |
-
@st.cache_resource
|
16 |
-
def
|
17 |
-
"""Loads the text generation pipeline."""
|
18 |
if not HF_TOKEN:
|
19 |
st.warning("HF_TOKEN secret not found. Ensure the model is public or add the token to secrets.")
|
20 |
-
# Decide if you want to stop or proceed cautiously
|
21 |
-
# st.stop() # Uncomment this line to halt execution if token is strictly required
|
22 |
|
23 |
try:
|
24 |
-
|
25 |
-
|
26 |
-
|
27 |
-
|
28 |
-
|
29 |
-
#
|
|
|
|
|
|
|
|
|
|
|
|
|
30 |
generator = pipeline(
|
31 |
"text-generation",
|
32 |
model=MODEL_NAME,
|
33 |
-
tokenizer=
|
34 |
-
torch_dtype
|
35 |
-
|
36 |
-
|
|
|
37 |
)
|
38 |
st.success(f"Model {MODEL_NAME} loaded successfully!")
|
39 |
-
return generator
|
40 |
except Exception as e:
|
41 |
-
st.error(f"Error loading model
|
42 |
-
st.error("
|
43 |
-
st.stop()
|
44 |
|
45 |
-
# --- Load
|
46 |
-
generator =
|
47 |
|
48 |
# --- Streamlit App UI ---
|
49 |
st.title("π° FinBuddy Assistant")
|
50 |
-
st.caption("
|
51 |
|
52 |
-
# Initialize chat history in session state if it doesn't exist
|
53 |
if "messages" not in st.session_state:
|
54 |
-
|
|
|
|
|
|
|
55 |
|
56 |
-
# Display past chat messages
|
57 |
for message in st.session_state.messages:
|
58 |
-
|
59 |
-
st.
|
|
|
60 |
|
61 |
-
# Get user input
|
62 |
if prompt := st.chat_input("Ask a question about finance..."):
|
63 |
-
# Add user
|
64 |
st.session_state.messages.append({"role": "user", "content": prompt})
|
65 |
with st.chat_message("user"):
|
66 |
st.markdown(prompt)
|
67 |
|
68 |
# Generate assistant response
|
69 |
with st.chat_message("assistant"):
|
70 |
-
message_placeholder = st.empty()
|
71 |
-
message_placeholder.markdown("Thinking...β³")
|
|
|
|
|
|
|
|
|
72 |
|
73 |
-
# ---
|
74 |
-
|
75 |
-
|
76 |
-
|
|
|
|
|
|
|
77 |
|
78 |
try:
|
79 |
# Generate response using the pipeline
|
80 |
outputs = generator(
|
81 |
-
|
82 |
-
max_new_tokens=512,
|
83 |
-
|
84 |
-
|
85 |
-
|
|
|
|
|
|
|
86 |
)
|
87 |
|
88 |
-
|
89 |
-
|
90 |
-
|
91 |
-
|
92 |
-
|
93 |
-
|
94 |
-
|
95 |
-
|
96 |
-
|
97 |
-
|
98 |
-
|
99 |
-
|
100 |
-
|
|
|
|
|
101 |
else:
|
102 |
-
|
|
|
103 |
|
104 |
-
# Handle cases where the response might be empty after cleaning
|
105 |
if not assistant_response:
|
106 |
-
|
107 |
-
|
108 |
else:
|
109 |
-
|
|
|
110 |
|
111 |
-
# Display the final response
|
112 |
message_placeholder.markdown(assistant_response)
|
113 |
-
# Add the final assistant response to session state
|
114 |
st.session_state.messages.append({"role": "assistant", "content": assistant_response})
|
115 |
|
116 |
except Exception as e:
|
117 |
error_message = f"Error during text generation: {e}"
|
118 |
st.error(error_message, icon="π₯")
|
119 |
-
message_placeholder.markdown("Sorry, an error occurred
|
120 |
-
# Add error indication to history
|
121 |
st.session_state.messages.append({"role": "assistant", "content": f"[Error: {e}]"})
|
|
|
1 |
import streamlit as st
|
2 |
import os
|
3 |
+
from transformers import pipeline, AutoTokenizer # Added AutoTokenizer
|
4 |
+
import torch
|
5 |
|
6 |
# --- Set Page Config FIRST ---
|
7 |
+
st.set_page_config(layout="wide")
|
8 |
|
9 |
# --- Configuration ---
|
10 |
+
# MODEL_NAME = "AdaptLLM/finance-LLM" # Old model
|
11 |
+
MODEL_NAME = "WiroAI/WiroAI-Finance-Qwen-1.5B" # New smaller model
|
12 |
HF_TOKEN = os.environ.get("HF_TOKEN")
|
13 |
|
14 |
# --- Model Loading (Cached by Streamlit for efficiency) ---
|
15 |
+
@st.cache_resource
|
16 |
+
def load_resources():
|
17 |
+
"""Loads the tokenizer and the text generation pipeline."""
|
18 |
if not HF_TOKEN:
|
19 |
st.warning("HF_TOKEN secret not found. Ensure the model is public or add the token to secrets.")
|
|
|
|
|
20 |
|
21 |
try:
|
22 |
+
st.info(f"Loading tokenizer for {MODEL_NAME}...")
|
23 |
+
tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME, use_auth_token=HF_TOKEN if HF_TOKEN else None)
|
24 |
+
st.success("Tokenizer loaded.")
|
25 |
+
|
26 |
+
# Determine device: Use GPU if available, otherwise CPU
|
27 |
+
# device_map="auto" might be problematic on CPU-only Spaces
|
28 |
+
# Start with device_map="auto", but fall back to explicit cpu if needed
|
29 |
+
device_map_setting = "auto"
|
30 |
+
# device = 0 if torch.cuda.is_available() else -1 # Alternative: explicit device
|
31 |
+
|
32 |
+
st.info(f"Loading model {MODEL_NAME}... (Using {device_map_setting}) This might take a while.")
|
33 |
+
# Use pipeline
|
34 |
generator = pipeline(
|
35 |
"text-generation",
|
36 |
model=MODEL_NAME,
|
37 |
+
tokenizer=tokenizer, # Pass loaded tokenizer
|
38 |
+
model_kwargs={"torch_dtype": torch.bfloat16}, # Use bfloat16 as per model card
|
39 |
+
device_map=device_map_setting,
|
40 |
+
# device=device # Use this if device_map causes issues
|
41 |
+
trust_remote_code=True
|
42 |
)
|
43 |
st.success(f"Model {MODEL_NAME} loaded successfully!")
|
44 |
+
return generator, tokenizer # Return both
|
45 |
except Exception as e:
|
46 |
+
st.error(f"Error loading model/tokenizer: {e}", icon="π₯")
|
47 |
+
st.error("Check memory limits, token access, or try removing device_map='auto'.")
|
48 |
+
st.stop()
|
49 |
|
50 |
+
# --- Load Resources ---
|
51 |
+
generator, tokenizer = load_resources()
|
52 |
|
53 |
# --- Streamlit App UI ---
|
54 |
st.title("π° FinBuddy Assistant")
|
55 |
+
st.caption(f"Model: {MODEL_NAME}")
|
56 |
|
|
|
57 |
if "messages" not in st.session_state:
|
58 |
+
# Add initial system message (as per model card example)
|
59 |
+
st.session_state.messages = [
|
60 |
+
{"role": "system", "content": "You are a finance chatbot developed by Wiro AI"}
|
61 |
+
]
|
62 |
|
63 |
+
# Display past chat messages (excluding system message)
|
64 |
for message in st.session_state.messages:
|
65 |
+
if message["role"] != "system": # Don't display system message
|
66 |
+
with st.chat_message(message["role"]):
|
67 |
+
st.markdown(message["content"])
|
68 |
|
69 |
+
# Get user input
|
70 |
if prompt := st.chat_input("Ask a question about finance..."):
|
71 |
+
# Add user prompt to state and display
|
72 |
st.session_state.messages.append({"role": "user", "content": prompt})
|
73 |
with st.chat_message("user"):
|
74 |
st.markdown(prompt)
|
75 |
|
76 |
# Generate assistant response
|
77 |
with st.chat_message("assistant"):
|
78 |
+
message_placeholder = st.empty()
|
79 |
+
message_placeholder.markdown("Thinking...β³")
|
80 |
+
|
81 |
+
# --- Prepare prompt for the model (use message history) ---
|
82 |
+
# Use the messages stored in session state (includes system prompt)
|
83 |
+
messages_for_api = st.session_state.messages
|
84 |
|
85 |
+
# --- Define terminators as per model card ---
|
86 |
+
terminators = [
|
87 |
+
tokenizer.eos_token_id,
|
88 |
+
tokenizer.convert_tokens_to_ids("<|end_of_text|>") # Qwen uses <|end_of_text|> usually
|
89 |
+
]
|
90 |
+
# Handle potential errors if the specific token doesn't exist
|
91 |
+
terminators = [term for term in terminators if term is not None and not isinstance(term, list)] # Filter out None or lists if conversion fails
|
92 |
|
93 |
try:
|
94 |
# Generate response using the pipeline
|
95 |
outputs = generator(
|
96 |
+
messages_for_api, # Pass the list of messages
|
97 |
+
max_new_tokens=512,
|
98 |
+
eos_token_id=terminators,
|
99 |
+
pad_token_id=tokenizer.eos_token_id, # Use EOS for padding
|
100 |
+
do_sample=True,
|
101 |
+
temperature=0.7, # Adjusted slightly from example
|
102 |
+
top_p=0.95, # Added common param
|
103 |
+
# top_k=50 # Optional parameter
|
104 |
)
|
105 |
|
106 |
+
# --- Extract response ---
|
107 |
+
# The output format is a list containing a dictionary with 'generated_text'
|
108 |
+
# which itself is a list of message dictionaries.
|
109 |
+
if (outputs and
|
110 |
+
isinstance(outputs, list) and
|
111 |
+
len(outputs) > 0 and
|
112 |
+
isinstance(outputs[0], dict) and
|
113 |
+
'generated_text' in outputs[0] and
|
114 |
+
isinstance(outputs[0]['generated_text'], list) and
|
115 |
+
len(outputs[0]['generated_text']) > 0):
|
116 |
+
|
117 |
+
# Get the last message dictionary in the generated list (should be the assistant's reply)
|
118 |
+
last_message = outputs[0]['generated_text'][-1]
|
119 |
+
if isinstance(last_message, dict) and last_message.get('role') == 'assistant':
|
120 |
+
assistant_response = last_message.get('content', "").strip()
|
121 |
else:
|
122 |
+
# Fallback if format is unexpected - try getting last element's text if it's a string?
|
123 |
+
assistant_response = str(outputs[0]['generated_text'][-1]).strip()
|
124 |
|
|
|
125 |
if not assistant_response:
|
126 |
+
assistant_response = "I generated an empty response."
|
127 |
+
|
128 |
else:
|
129 |
+
print("Unexpected output format:", outputs) # Log for debugging
|
130 |
+
assistant_response = "Sorry, I couldn't parse the response format."
|
131 |
|
|
|
132 |
message_placeholder.markdown(assistant_response)
|
|
|
133 |
st.session_state.messages.append({"role": "assistant", "content": assistant_response})
|
134 |
|
135 |
except Exception as e:
|
136 |
error_message = f"Error during text generation: {e}"
|
137 |
st.error(error_message, icon="π₯")
|
138 |
+
message_placeholder.markdown("Sorry, an error occurred generating the response.")
|
|
|
139 |
st.session_state.messages.append({"role": "assistant", "content": f"[Error: {e}]"})
|