File size: 4,400 Bytes
3ab546a 460ac99 3ab546a e2f1245 3ab546a 9b89cea 3ab546a 9ad2d00 3ab546a 10d4eb0 3d6dae7 3ab546a 460ac99 3ab546a 9ad2d00 3ab546a 51feb0f 3ab546a |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 |
import os
os.system('pip install pyyaml==5.1')
# workaround: install old version of pytorch since detectron2 hasn't released packages for pytorch 1.9 (issue: https://github.com/facebookresearch/detectron2/issues/3158)
os.system('pip install torch==1.8.0+cu101 torchvision==0.9.0+cu101 -f https://download.pytorch.org/whl/torch_stable.html')
# install detectron2 that matches pytorch 1.8
# See https://detectron2.readthedocs.io/tutorials/install.html for instructions
os.system('pip install -q detectron2 -f https://dl.fbaipublicfiles.com/detectron2/wheels/cu101/torch1.8/index.html')
## install PyTesseract
os.system('pip install -q pytesseract')
import gradio as gr
import numpy as np
from transformers import LayoutLMv2Processor, LayoutLMv2ForTokenClassification
from datasets import load_dataset
from PIL import Image, ImageDraw, ImageFont
processor = LayoutLMv2Processor.from_pretrained("microsoft/layoutlmv2-base-uncased")
model = LayoutLMv2ForTokenClassification.from_pretrained("nielsr/layoutlmv2-finetuned-funsd")
#model = LayoutLMv2ForTokenClassification.from_pretrained("mishtert/iec")
# load image example
dataset = load_dataset("nielsr/funsd", split="test")
#dataset = load_dataset("mishtert/niefunsd", split="test")
image = Image.open(dataset[0]["image_path"]).convert("RGB")
image = Image.open("./invoice.png")
image.save("document.png")
Image.open("./invoice2.png").convert("RGB").save("document1.png")
Image.open("./invoice3.png").convert("RGB").save("document2.png")
# define id2label, label2color
labels = dataset.features['ner_tags'].feature.names
id2label = {v: k for v, k in enumerate(labels)}
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
title = "Invoice Extraction & Categorization"
description = "Invoice text identified (extraction) and categorized."
#examples =[['document.png']]
examples =[['document.png'],['invoice2.png'],['invoice3.png']]
css = ".output-image, .input-image {height: 40rem !important; width: 100% !important;}"
#css = "@media screen and (max-width: 600px) { .output_image, .input_image {height:20rem !important; width: 100% !important;} }"
# css = ".output_image, .input_image {height: 600px !important}"
css = ".image-preview {height: auto !important;}"
iface = gr.Interface(fn=process_image,
inputs=gr.inputs.Image(type="pil"),
outputs= gr.outputs.Image(type="pil", label="Identified & Categorized Image"),
# outputs= [gr.outputs.Image(type="pil", label="Identified & Categorized Image"),
# gr.outputs.Textbox(type="text", label="Identified & Categorized Image")],
title=title,
description=description,
# article=article,
examples=examples,
css=css,
enable_queue=True)
iface.launch(debug=True) |