Spaces:
Sleeping
Sleeping
Create app.py
Browse files
app.py
ADDED
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import streamlit as st
|
2 |
+
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
|
3 |
+
|
4 |
+
# Step 1: Load the fine-tuned model and tokenizer
|
5 |
+
MODEL_NAME = "ridahabbash/AccR4" # Replace with your model's Hub ID
|
6 |
+
tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
|
7 |
+
model = AutoModelForSeq2SeqLM.from_pretrained(MODEL_NAME)
|
8 |
+
|
9 |
+
# Step 2: Define the prediction function
|
10 |
+
def generate_report(prompt):
|
11 |
+
# Tokenize input
|
12 |
+
inputs = tokenizer(prompt, return_tensors="pt", max_length=512, truncation=True)
|
13 |
+
|
14 |
+
# Generate output
|
15 |
+
outputs = model.generate(**inputs, max_length=128)
|
16 |
+
|
17 |
+
# Decode and return the result
|
18 |
+
report = tokenizer.decode(outputs[0], skip_special_tokens=True)
|
19 |
+
return report
|
20 |
+
|
21 |
+
# Step 3: Create the Streamlit interface
|
22 |
+
st.title("Accounting Report Generator")
|
23 |
+
st.markdown("Enter a ledger entry below, and the model will generate a report.")
|
24 |
+
|
25 |
+
# Input textbox
|
26 |
+
prompt = st.text_area("Ledger Entry", placeholder="Enter ledger details here...")
|
27 |
+
|
28 |
+
# Generate button
|
29 |
+
if st.button("Generate Report"):
|
30 |
+
if prompt.strip() == "":
|
31 |
+
st.error("Please enter a valid ledger entry.")
|
32 |
+
else:
|
33 |
+
# Generate report
|
34 |
+
with st.spinner("Generating report..."):
|
35 |
+
report = generate_report(prompt)
|
36 |
+
# Display the report
|
37 |
+
st.success("Report generated successfully!")
|
38 |
+
st.subheader("Generated Report:")
|
39 |
+
st.write(report)
|
40 |
+
|
41 |
+
|