phamvanhanh6720 / idcardvnrecognition

Recognize and extract information from ID Card VietNam

IDCardVNRecognition

• Recommend: GPU + OS(Ubuntu)

Update

• I recently updated the new version that is easy to install and is improved slightly in accuracy and performance.

Install Dependencies

# Install an ASGI server, for production such as Uvicorn or Hypercorn.
pip3 install uvicorn[standard]

pip3 install -r requirements.txt

Run Server

uvicorn run:app --host='hostname'

idcardvnrecognition's Issues

Lỗi khi chạy Demo

Em tải về chạy thử thì nó ra lỗi này. Em xin cách khắc phục ạ!

D:\Downloads\IDCardVNRecognition-master>python run.py
ENV is set to: production
2020-10-13 10:42:50.586144: I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library cudart64_101.dll
File exists: C:\Users\VBD.cache/gdown\https-COLON--SLASH--SLASH-drive.google.com-SLASH-uc-QUESTION-export-EQUAL-download&id-EQUAL-1-olev206xLgXYf7rnwHrcZLxxLg5rs0p
Traceback (most recent call last):
File "run.py", line 1, in
from app import app
File "D:\Downloads\IDCardVNRecognition-master\app_init_.py", line 10, in
from app import main
File "D:\Downloads\IDCardVNRecognition-master\app\main.py", line 33, in
reader = Predictor(config)
File "D:\Downloads\IDCardVNRecognition-master\reader\reader.py", line 22, in init
model.load_state_dict(torch.load(weights, map_location=torch.device(device)))
File "C:\Users\VBD\AppData\Roaming\Python\Python37\site-packages\torch\nn\modules\module.py", line 1045, in load_state_dict
self.class.name, "\n\t".join(error_msgs)))
RuntimeError: Error(s) in loading state_dict for VietOCR:
size mismatch for transformer.embed_tgt.weight: copying a param with shape torch.Size([232, 256]) from checkpoint, the shape in current model is torch.Size([233, 256]).
size mismatch for transformer.fc.weight: copying a param with shape torch.Size([232, 256]) from checkpoint, the shape in current model is torch.Size([233, 256]).
size mismatch for transformer.fc.bias: copying a param with shape torch.Size([232]) from checkpoint, the shape in current model is torch.Size([233]).

Error: requests.exceptions.MissingSchema: Invalid URL '': No schema supplied. Perhaps you meant http://?

Fixed:
uninstall older vietocr: python3 -m pip uninstall vietocr
reinstall newest vietocr: python3 -m pip install vietocr

Mấy file này ở đây vậy bạn

hướng dẫn em fix lỗi này với ạ

Anh có thể cho em xin file hoặc đường dẫn đến cropper với detector được không ạ. Em cảm ơn ạ.
Traceback (most recent call last):
File "run.py", line 1, in
from app.main import app
File "D:\project\v4\IDCardVNRecognition\app\main.py", line 31, in
cropper = Cropper(config_path=download_weights(cfg['cropper']['cfg']),
KeyError: 'cropper'

How to download config and weight?

Dear author, I recently used this code to do some work, but when running main.py, I always get a requests connection error, I found that this error is appearing when the model configuration and weights are downloaded, what should I do?

any help would be appreciate.

Lỗi khi chạy demo

em cài đầy đủ thư viện và chạy demo thì xuất hiện lỗi này, anh cho em cách khắc phục được k ạ. Em cảm ơn ạ

hieu@hieu-X510UQR:~/Desktop/IDCardVNRecognition$ python3 run.py
ENV is set to: production
2020-10-18 19:02:29.610328: I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library libcudart.so.10.1
Traceback (most recent call last):
File "run.py", line 1, in
from app import app
File "/home/hieu/Desktop/IDCardVNRecognition/app/init.py", line 10, in
from app import main
File "/home/hieu/Desktop/IDCardVNRecognition/app/main.py", line 33, in
reader = Predictor(config)
File "/home/hieu/Desktop/IDCardVNRecognition/reader/reader.py", line 14, in init
model, vocab = build_model(config)
File "/home/hieu/.local/lib/python3.6/site-packages/vietocr/tool/translate.py", line 111, in build_model
config['transformer'])
File "/home/hieu/.local/lib/python3.6/site-packages/vietocr/model/transformerocr.py", line 13, in init
self.cnn = CNN(backbone, **cnn_args)
File "/home/hieu/.local/lib/python3.6/site-packages/vietocr/model/cnn.py", line 15, in init
self.model = vgg.vgg19_bn(**kwargs)
TypeError: vgg19_bn() got an unexpected keyword argument 'pretrained'

Tìm input, output node name

Chào bạn, bạn có thể cho mình hỏi 1 câu hơi ngoài lề một chút: Làm sao để tìm ra input và output node name khi gọi tf serving ?

Ví dụ như trong repo của bạn có sử dụng request.inputs["input_1"] và result = result.outputs["tf_op_layer_concat_14"].float_val

Mình có thử dùng tool Netron nhưng chỉ tìm được input.

Cám ơn các bạn rất nhiều.