Spaces:
Runtime error
Runtime error
Commit
·
3579971
1
Parent(s):
97f9f1f
back to final state
Browse files
app.py
CHANGED
@@ -1,6 +1,5 @@
|
|
1 |
-
import gradio as gr
|
2 |
-
import csv
|
3 |
import re
|
|
|
4 |
|
5 |
import torch
|
6 |
from transformers import DonutProcessor, VisionEncoderDecoderModel
|
@@ -38,37 +37,20 @@ def process_document(image):
|
|
38 |
sequence = sequence.replace(processor.tokenizer.eos_token, "").replace(processor.tokenizer.pad_token, "")
|
39 |
sequence = re.sub(r"<.*?>", "", sequence, count=1).strip() # remove first task start token
|
40 |
|
41 |
-
with open('output.csv', 'a', newline='') as file:
|
42 |
-
writer = csv.writer(file)
|
43 |
-
writer.writerow(["image", sequence])
|
44 |
-
|
45 |
return processor.token2json(sequence)
|
46 |
|
47 |
-
def download_csv():
|
48 |
-
return "output.csv"
|
49 |
-
|
50 |
description = "To use it, simply upload your image and click 'submit', or click one of the examples to load them. Read more at the links below."
|
51 |
article = "<p style='text-align: center'><a href='https://arxiv.org/abs/2111.15664' target='_blank'>Donut: OCR-free Document Understanding Transformer</a> | <a href='https://github.com/clovaai/donut' target='_blank'>Github Repo</a></p>"
|
52 |
|
53 |
-
|
54 |
-
|
55 |
-
|
56 |
-
|
57 |
-
|
58 |
-
download_button = gr.outputs.Button("Download CSV")
|
59 |
-
|
60 |
-
interface = gr.Interface(
|
61 |
-
process_document,
|
62 |
-
inputs=inputs,
|
63 |
-
outputs=outputs,
|
64 |
title="Donut 🍩 for Document Parsing and Converting the Image to Database",
|
65 |
description=description,
|
66 |
article=article,
|
67 |
-
examples=examples,
|
68 |
enable_queue=True,
|
69 |
-
|
70 |
-
|
71 |
-
download_button=download_button
|
72 |
-
)
|
73 |
|
74 |
-
|
|
|
|
|
|
|
1 |
import re
|
2 |
+
import gradio as gr
|
3 |
|
4 |
import torch
|
5 |
from transformers import DonutProcessor, VisionEncoderDecoderModel
|
|
|
37 |
sequence = sequence.replace(processor.tokenizer.eos_token, "").replace(processor.tokenizer.pad_token, "")
|
38 |
sequence = re.sub(r"<.*?>", "", sequence, count=1).strip() # remove first task start token
|
39 |
|
|
|
|
|
|
|
|
|
40 |
return processor.token2json(sequence)
|
41 |
|
|
|
|
|
|
|
42 |
description = "To use it, simply upload your image and click 'submit', or click one of the examples to load them. Read more at the links below."
|
43 |
article = "<p style='text-align: center'><a href='https://arxiv.org/abs/2111.15664' target='_blank'>Donut: OCR-free Document Understanding Transformer</a> | <a href='https://github.com/clovaai/donut' target='_blank'>Github Repo</a></p>"
|
44 |
|
45 |
+
demo = gr.Interface(
|
46 |
+
fn=process_document,
|
47 |
+
inputs="image",
|
48 |
+
outputs="json",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
49 |
title="Donut 🍩 for Document Parsing and Converting the Image to Database",
|
50 |
description=description,
|
51 |
article=article,
|
|
|
52 |
enable_queue=True,
|
53 |
+
examples=[["example.png"], ["example_2.png"], ["example_3.png"]],
|
54 |
+
cache_examples=False)
|
|
|
|
|
55 |
|
56 |
+
demo.launch()
|