add app
Browse files- app.py +91 -0
- requirements.txt +4 -0
app.py
ADDED
@@ -0,0 +1,91 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import gradio as gr
|
2 |
+
import pandas as pd
|
3 |
+
from transformers import TapexTokenizer, BartForConditionalGeneration, pipeline
|
4 |
+
|
5 |
+
# Initialize TAPEX (Microsoft) model and tokenizer
|
6 |
+
tokenizer_tapex = TapexTokenizer.from_pretrained("microsoft/tapex-large-finetuned-wtq")
|
7 |
+
model_tapex = BartForConditionalGeneration.from_pretrained("microsoft/tapex-large-finetuned-wtq")
|
8 |
+
|
9 |
+
# Initialize TAPAS (Google) models and pipelines
|
10 |
+
pipe_tapas = pipeline(task="table-question-answering", model="google/tapas-large-finetuned-wtq")
|
11 |
+
pipe_tapas2 = pipeline(task="table-question-answering", model="google/tapas-large-finetuned-wikisql-supervised")
|
12 |
+
|
13 |
+
def process_table_query(query, table_data):
|
14 |
+
"""
|
15 |
+
Process a query and CSV data using TAPEX.
|
16 |
+
"""
|
17 |
+
# Convert all columns in the table to strings for TAPEX compatibility
|
18 |
+
table_data = table_data.astype(str)
|
19 |
+
|
20 |
+
# Microsoft TAPEX model (using TAPEX tokenizer and model)
|
21 |
+
encoding = tokenizer_tapex(table=table_data, query=query, return_tensors="pt", max_length=1024, truncation=True)
|
22 |
+
outputs = model_tapex.generate(**encoding)
|
23 |
+
result_tapex = tokenizer_tapex.batch_decode(outputs, skip_special_tokens=True)[0]
|
24 |
+
|
25 |
+
return result_tapex
|
26 |
+
|
27 |
+
# Gradio interface
|
28 |
+
def answer_query_from_csv(query, file):
|
29 |
+
"""
|
30 |
+
Function to handle file input and return model results.
|
31 |
+
"""
|
32 |
+
# Read the file into a DataFrame
|
33 |
+
table_data = pd.read_csv(file)
|
34 |
+
|
35 |
+
# Convert object-type columns to lowercase (if they are valid strings)
|
36 |
+
for column in table_data.columns:
|
37 |
+
if table_data[column].dtype == 'object':
|
38 |
+
table_data[column] = table_data[column].apply(lambda x: x.lower() if isinstance(x, str) else x)
|
39 |
+
|
40 |
+
# Convert all table cells to strings for TAPEX compatibility
|
41 |
+
table_data = table_data.astype(str)
|
42 |
+
|
43 |
+
# Extract year, month, day, and time components for datetime columns
|
44 |
+
for column in table_data.columns:
|
45 |
+
if pd.api.types.is_datetime64_any_dtype(table_data[column]):
|
46 |
+
table_data[f'{column}_year'] = table_data[column].dt.year
|
47 |
+
table_data[f'{column}_month'] = table_data[column].dt.month
|
48 |
+
table_data[f'{column}_day'] = table_data[column].dt.day
|
49 |
+
table_data[f'{column}_time'] = table_data[column].dt.strftime('%H:%M:%S')
|
50 |
+
|
51 |
+
# Process the CSV file and query
|
52 |
+
result_tapex = process_table_query(query, table_data)
|
53 |
+
|
54 |
+
# Process the query using TAPAS pipelines
|
55 |
+
result_tapas = pipe_tapas(table=table_data, query=query)['cells'][0]
|
56 |
+
result_tapas2 = pipe_tapas2(table=table_data, query=query)['cells'][0]
|
57 |
+
|
58 |
+
return result_tapex, result_tapas, result_tapas2
|
59 |
+
|
60 |
+
# Create Gradio interface
|
61 |
+
with gr.Blocks() as interface:
|
62 |
+
gr.Markdown("# Table Question Answering with TAPEX and TAPAS Models")
|
63 |
+
|
64 |
+
# Add a notice about the token limit
|
65 |
+
gr.Markdown("### Note: Only the first 1024 tokens (query + table data) will be considered. If your table is too large, it will be truncated to fit within this limit.")
|
66 |
+
|
67 |
+
# Two-column layout (input on the left, output on the right)
|
68 |
+
with gr.Row():
|
69 |
+
with gr.Column():
|
70 |
+
# Input fields for the query and file
|
71 |
+
query_input = gr.Textbox(label="Enter your query:")
|
72 |
+
csv_input = gr.File(label="Upload your CSV file")
|
73 |
+
|
74 |
+
with gr.Column():
|
75 |
+
# Output textboxes for the answers
|
76 |
+
result_tapex = gr.Textbox(label="TAPEX Answer")
|
77 |
+
result_tapas = gr.Textbox(label="TAPAS (WikiTableQuestions) Answer")
|
78 |
+
result_tapas2 = gr.Textbox(label="TAPAS (WikiSQL) Answer")
|
79 |
+
|
80 |
+
# Submit button
|
81 |
+
submit_btn = gr.Button("Submit")
|
82 |
+
|
83 |
+
# Action when submit button is clicked
|
84 |
+
submit_btn.click(
|
85 |
+
fn=answer_query_from_csv,
|
86 |
+
inputs=[query_input, csv_input],
|
87 |
+
outputs=[result_tapex, result_tapas, result_tapas2]
|
88 |
+
)
|
89 |
+
|
90 |
+
# Launch the Gradio interface
|
91 |
+
interface.launch(share=True)
|
requirements.txt
ADDED
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
1 |
+
gradio
|
2 |
+
pandas
|
3 |
+
transformers
|
4 |
+
torch
|