layoutlmv2-base-uncased/app.py

82 lines
3.0 KiB
Python

import gradio as gr
import numpy as np
from transformers import LayoutLMv2Processor, LayoutLMv2ForTokenClassification
from datasets import load_dataset
from PIL import Image, ImageDraw, ImageFont
from gradio.themes.utils import sizes
theme = gr.themes.Default(radius_size=sizes.radius_none).set(
block_label_text_color = '#4D63FF',
block_title_text_color = '#4D63FF',
button_primary_text_color = '#4D63FF',
button_primary_background_fill='#FFFFFF',
button_primary_border_color='#4D63FF',
button_primary_background_fill_hover='#EDEFFF',
)
processor = LayoutLMv2Processor.from_pretrained("microsoft/layoutlmv2-base-uncased")
model = LayoutLMv2ForTokenClassification.from_pretrained("nielsr/layoutlmv2-finetuned-funsd")
labels = ['O', 'B-HEADER', 'I-HEADER', 'B-QUESTION', 'I-QUESTION', 'B-ANSWER', 'I-ANSWER']
id2label = {0: 'O', 1: 'B-HEADER', 2: 'I-HEADER', 3: 'B-QUESTION', 4: 'I-QUESTION', 5: 'B-ANSWER', 6: 'I-ANSWER'}
label2color = {'question':'blue', 'answer':'green', 'header':'orange', 'other':'violet'}
def unnormalize_box(bbox, width, height):
return [
width * (bbox[0] / 1000),
height * (bbox[1] / 1000),
width * (bbox[2] / 1000),
height * (bbox[3] / 1000),
]
def iob_to_label(label):
label = label[2:]
if not label:
return 'other'
return label
def process_image(image):
width, height = image.size
# encode
encoding = processor(image, truncation=True, return_offsets_mapping=True, return_tensors="pt")
offset_mapping = encoding.pop('offset_mapping')
# forward pass
outputs = model(**encoding)
# get predictions
predictions = outputs.logits.argmax(-1).squeeze().tolist()
token_boxes = encoding.bbox.squeeze().tolist()
# only keep non-subword predictions
is_subword = np.array(offset_mapping.squeeze().tolist())[:,0] != 0
true_predictions = [id2label[pred] for idx, pred in enumerate(predictions) if not is_subword[idx]]
true_boxes = [unnormalize_box(box, width, height) for idx, box in enumerate(token_boxes) if not is_subword[idx]]
# draw predictions over the image
draw = ImageDraw.Draw(image)
font = ImageFont.load_default()
for prediction, box in zip(true_predictions, true_boxes):
predicted_label = iob_to_label(prediction).lower()
draw.rectangle(box, outline=label2color[predicted_label])
draw.text((box[0]+10, box[1]-10), text=predicted_label, fill=label2color[predicted_label], font=font)
return image
with gr.Blocks(theme=theme, css="footer {visibility: hidden}") as demo:
with gr.Row():
with gr.Column():
image_input =gr.inputs.Image(type="pil", label="图片")
with gr.Row():
button = gr.Button("提交", variant="primary")
image_output = gr.Image(label="图片")
button.click(fn=process_image, inputs=image_input, outputs=image_output)
examples = gr.Examples(examples=[['document.png']], inputs=[image_input], label="例子")
if __name__ == "__main__":
demo.queue(concurrency_count=3).launch(server_name = "0.0.0.0")