Spaces:
Running
Running
Update app.py
Browse files
app.py
CHANGED
@@ -11,11 +11,12 @@ from transformers import pipeline
|
|
11 |
import gradio as gr
|
12 |
from fastapi.responses import RedirectResponse
|
13 |
import numpy as np
|
|
|
14 |
# Initialize FastAPI
|
15 |
app = FastAPI()
|
16 |
|
17 |
-
# Load AI Model for Question Answering
|
18 |
-
qa_pipeline = pipeline("
|
19 |
|
20 |
# Load Pretrained Object Detection Model (Torchvision)
|
21 |
model = fasterrcnn_resnet50_fpn(pretrained=True)
|
@@ -60,7 +61,7 @@ def extract_text_from_excel(excel_file):
|
|
60 |
text.append(" ".join(map(str, row)))
|
61 |
return "\n".join(text)
|
62 |
|
63 |
-
# Function to
|
64 |
def extract_text_from_image(image_file):
|
65 |
if isinstance(image_file, np.ndarray): # Check if input is a NumPy array
|
66 |
image = Image.fromarray(image_file) # Convert NumPy array to PIL image
|
@@ -70,7 +71,8 @@ def extract_text_from_image(image_file):
|
|
70 |
reader = easyocr.Reader(["en"])
|
71 |
result = reader.readtext(np.array(image)) # Convert PIL image back to NumPy array
|
72 |
return " ".join([res[1] for res in result])
|
73 |
-
|
|
|
74 |
def answer_question_from_document(file, question):
|
75 |
file_ext = file.name.split(".")[-1].lower()
|
76 |
|
@@ -89,10 +91,10 @@ def answer_question_from_document(file, question):
|
|
89 |
return "No text extracted from the document."
|
90 |
|
91 |
truncated_text = truncate_text(text)
|
92 |
-
input_text = f"
|
93 |
-
response = qa_pipeline(input_text)
|
94 |
|
95 |
-
return response[0]["
|
96 |
|
97 |
# Function to answer questions based on image content
|
98 |
def answer_question_from_image(image, question):
|
@@ -101,10 +103,10 @@ def answer_question_from_image(image, question):
|
|
101 |
return "No meaningful content detected in the image."
|
102 |
|
103 |
truncated_text = truncate_text(image_text)
|
104 |
-
input_text = f"
|
105 |
-
response = qa_pipeline(input_text)
|
106 |
|
107 |
-
return response[0]["
|
108 |
|
109 |
# Gradio UI for Document & Image QA
|
110 |
doc_interface = gr.Interface(
|
|
|
11 |
import gradio as gr
|
12 |
from fastapi.responses import RedirectResponse
|
13 |
import numpy as np
|
14 |
+
|
15 |
# Initialize FastAPI
|
16 |
app = FastAPI()
|
17 |
|
18 |
+
# Load AI Model for Question Answering (Summarization-based approach)
|
19 |
+
qa_pipeline = pipeline("summarization", model="facebook/bart-large-cnn", tokenizer="facebook/bart-large-cnn")
|
20 |
|
21 |
# Load Pretrained Object Detection Model (Torchvision)
|
22 |
model = fasterrcnn_resnet50_fpn(pretrained=True)
|
|
|
61 |
text.append(" ".join(map(str, row)))
|
62 |
return "\n".join(text)
|
63 |
|
64 |
+
# Function to extract text from image
|
65 |
def extract_text_from_image(image_file):
|
66 |
if isinstance(image_file, np.ndarray): # Check if input is a NumPy array
|
67 |
image = Image.fromarray(image_file) # Convert NumPy array to PIL image
|
|
|
71 |
reader = easyocr.Reader(["en"])
|
72 |
result = reader.readtext(np.array(image)) # Convert PIL image back to NumPy array
|
73 |
return " ".join([res[1] for res in result])
|
74 |
+
|
75 |
+
# Function to answer questions based on document content using BART summarization
|
76 |
def answer_question_from_document(file, question):
|
77 |
file_ext = file.name.split(".")[-1].lower()
|
78 |
|
|
|
91 |
return "No text extracted from the document."
|
92 |
|
93 |
truncated_text = truncate_text(text)
|
94 |
+
input_text = f"Context: {truncated_text} Question: {question}"
|
95 |
+
response = qa_pipeline(input_text, max_length=100, min_length=30, do_sample=False)
|
96 |
|
97 |
+
return response[0]["summary_text"]
|
98 |
|
99 |
# Function to answer questions based on image content
|
100 |
def answer_question_from_image(image, question):
|
|
|
103 |
return "No meaningful content detected in the image."
|
104 |
|
105 |
truncated_text = truncate_text(image_text)
|
106 |
+
input_text = f"Context: {truncated_text} Question: {question}"
|
107 |
+
response = qa_pipeline(input_text, max_length=100, min_length=30, do_sample=False)
|
108 |
|
109 |
+
return response[0]["summary_text"]
|
110 |
|
111 |
# Gradio UI for Document & Image QA
|
112 |
doc_interface = gr.Interface(
|