a micro OCR network with 0.07mb params.

william

Last update: Aug 6, 2022

Related tags

Computer Vision MicroOCR

Overview

MicroOCR

a micro OCR network with 0.07mb params.

    Layer (type)               Output Shape         Param #

        Conv2d-1            [-1, 64, 8, 32]           3,136
   BatchNorm2d-2            [-1, 64, 8, 32]             128
          GELU-3            [-1, 64, 8, 32]               0
     ConvBNACT-4            [-1, 64, 8, 32]               0
        Conv2d-5            [-1, 64, 8, 32]             640
   BatchNorm2d-6            [-1, 64, 8, 32]             128
          GELU-7            [-1, 64, 8, 32]               0
     ConvBNACT-8            [-1, 64, 8, 32]               0
        Conv2d-9            [-1, 64, 8, 32]           4,160
  BatchNorm2d-10            [-1, 64, 8, 32]             128
         GELU-11            [-1, 64, 8, 32]               0
    ConvBNACT-12            [-1, 64, 8, 32]               0
   MicroBlock-13            [-1, 64, 8, 32]               0
       Conv2d-14            [-1, 64, 8, 32]             640
  BatchNorm2d-15            [-1, 64, 8, 32]             128
         GELU-16            [-1, 64, 8, 32]               0
    ConvBNACT-17            [-1, 64, 8, 32]               0
       Conv2d-18            [-1, 64, 8, 32]           4,160
  BatchNorm2d-19            [-1, 64, 8, 32]             128
         GELU-20            [-1, 64, 8, 32]               0
    ConvBNACT-21            [-1, 64, 8, 32]               0
   MicroBlock-22            [-1, 64, 8, 32]               0
      Flatten-23              [-1, 64, 256]               0
AdaptiveAvgPool1d-24           [-1, 64, 30]               0
       Linear-25               [-1, 30, 60]           3,900

Total params: 17,276
Trainable params: 17,276
Non-trainable params: 0
Input size (MB): 0.05
Forward/backward pass size (MB): 2.90
Params size (MB): 0.07
Estimated Total Size (MB): 3.02

Script Description

MicroOCR
├── README.md                                   # Descriptions about MicroNet
├── collatefn.py                                # collatefn
├── ctc_label_converter.py                      # accuracy metric for MicroNet
├── dataset.py                                  # Data preprocessing for training and evaluation
├── demo.py                                     # demo
├── gen_image.py                                # generate image for train and eval
├── infer_tool.py                               # inference tool
├── keys.py                                     # character
├── loss.py                                     # Ctcloss definition
├── metric.py                                   # accuracy metric for MicroNet
├── model.py                                    # MicroNet
├── train.py                                    # train the model

Generate data for train and eval

python gen_image.py

Training

python train.py

Inference

python demo.py

You might also like...

A pure pytorch implemented ocr project including text detection and recognition

ocr.pytorch A pure pytorch implemented ocr project. Text detection is based CTPN and text recognition is based CRNN. More detection and recognition me

444 Dec 30, 2022

python ocr using tesseract/ with EAST opencv detector

pytextractor python ocr using tesseract/ with EAST opencv text detector Uses the EAST opencv detector defined here with pytesseract to extract text(de

38 Dec 5, 2022

Run tesseract with the tesserocr bindings with @OCR-D's interfaces

ocrd_tesserocr Crop, deskew, segment into regions / tables / lines / words, or recognize with tesserocr Introduction This package offers OCR-D complia

38 Oct 14, 2022

A set of workflows for corpus building through OCR, post-correction and normalisation

PICCL: Philosophical Integrator of Computational and Corpus Libraries PICCL offers a workflow for corpus building and builds on a variety of tools. Th

41 Dec 27, 2022

Tensorflow-based CNN+LSTM trained with CTC-loss for OCR

Overview This collection demonstrates how to construct and train a deep, bidirectional stacked LSTM using CNN features as input with CTC loss to perfo

489 Dec 21, 2022

🖺 OCR using tensorflow with attention

tensorflow-ocr 🖺 OCR using tensorflow with attention, batteries included Installation git clone --recursive http://github.com/pannous/tensorflow-ocr

646 Nov 11, 2022

A tool for extracting text from scanned documents (via OCR), with user-defined post-processing.

The project is based on older versions of tesseract and other tools, and is now superseded by another project which allows for more granular control o

32 Jul 24, 2022

MXNet OCR implementation. Including text recognition and detection.

insightocr Text Recognition Accuracy on Chinese dataset by caffe-ocr Network LSTM 4x1 Pooling Gray Test Acc SimpleNet N Y Y 99.37% SE-ResNet34 N Y Y 9

99 Nov 1, 2022

CNN+LSTM+CTC based OCR implemented using tensorflow.

CNN_LSTM_CTC_Tensorflow CNN+LSTM+CTC based OCR(Optical Character Recognition) implemented using tensorflow. Note: there is No restriction on the numbe

356 Dec 8, 2022

Comments

ZeroDivisionError: float division by zero

(env) $ python train.py 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 5000/5000 [00:00<00:00, 78430.46it/s] 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 500/500 [00:00<00:00, 79000.68it/s] 0.0 0.0 Traceback (most recent call last): File "train.py", line 255, in main() File "train.py", line 251, in main train_model(cfg) File "train.py", line 131, in train_model model, device, test_loader, converter, metric, loss_func, cfg.show_str_size) File "train.py", line 62, in test_model return running_word_corrects / running_all, running_char_corrects / running_all_char_correct, ZeroDivisionError: float division by zero

opened by yisampi 1