Spaces:
Running
Running
Update app.py
Browse files
app.py
CHANGED
@@ -105,7 +105,6 @@ def process_image(input_image):
|
|
105 |
output = ocr_model.generate(
|
106 |
**inputs,
|
107 |
max_new_tokens=40,
|
108 |
-
temperature=0.2,
|
109 |
num_beams=1,
|
110 |
do_sample=False # Greedy decoding
|
111 |
)
|
@@ -138,11 +137,9 @@ iface = gr.Interface(
|
|
138 |
gr.Image(type="pil", label="Cropped & Preprocessed Image"),
|
139 |
gr.Textbox(label="Extracted Text")
|
140 |
],
|
141 |
-
title="Document OCR with YOLO
|
142 |
description=(
|
143 |
-
"Upload an image of a document. The app enhances the image,
|
144 |
-
"to detect and crop the document (front/back) with a confidence threshold of 0.85, and "
|
145 |
-
"then extracts text using an OCR model with a corresponding prompt."
|
146 |
),
|
147 |
allow_flagging="never" # Disable flagging to simplify UI
|
148 |
)
|
|
|
105 |
output = ocr_model.generate(
|
106 |
**inputs,
|
107 |
max_new_tokens=40,
|
|
|
108 |
num_beams=1,
|
109 |
do_sample=False # Greedy decoding
|
110 |
)
|
|
|
137 |
gr.Image(type="pil", label="Cropped & Preprocessed Image"),
|
138 |
gr.Textbox(label="Extracted Text")
|
139 |
],
|
140 |
+
title="Document OCR with YOLO and OLMOCR",
|
141 |
description=(
|
142 |
+
"Upload an image of a document. The app enhances the image, then extracts text using an OCR model."
|
|
|
|
|
143 |
),
|
144 |
allow_flagging="never" # Disable flagging to simplify UI
|
145 |
)
|