Vietnamese Handwriting Recognition

Dataset

• Vietnamese Handwriting Dataset: https://www.dropbox.com/s/zk82qkidkqma8bj/vn_handwriting_data.zip?dl=0

Command

export PYTHONPATH=path-to-project

cp src/example_config.py src/config.py

Result

• Model trained with 25 epochs: https://github.com/huyhoang17/Colab_Temporary/blob/master/Training_CTC_Vietnamese_Recognition_25epochs.ipynb

• [Overfitting model] Model trained with 10 epochs (initial weights from 25 epochs), reduced LR from 0.02 to 0.005: https://github.com/huyhoang17/Colab_Temporary/blob/master/%5BOverfitting%5DTraining_CTC_Vietnamese_Recognition_40epochs.ipynb

TODO

• Add Dockerfile
• Deploy simple demo with Tensorflow Serving
• Simple API
• Refactor code
• Add Visual Callback :heavy_check_mark:
  • callbacks.py
• Train code with Colab's TPU?
• Convert to Pytorch code

Prevent Overfitting

• Data Augmentation :heavy_check_mark:
  • data_utils.py and datagen.py
• Generate more data
  • https://arxiv.org/abs/1308.0850
  • https://arxiv.org/abs/1801.08624
• Simplifly the model
• Early Stopping
• Cross Validation
• Dropout (NN)
• Use Transfer Learing!
• ...

Reference

Colab Notebook:

• https://github.com/huyhoang17/Colab_Temporary/blob/master/Training_CTC_Vietnamese_Recognition_10epochs.ipynb

Papers

• https://arxiv.org/pdf/1804.01527.pdf
• https://arxiv.org/abs/1801.08624
• https://arxiv.org/abs/1308.0850
  • Blog: https://greydanus.github.io/2016/08/21/handwriting/
  • https://github.com/tensorflow/magenta/blob/master/magenta/reviews/summary_generation_sequences.md
  • https://github.com/sjvasquez/handwriting-synthesis
  • https://github.com/greydanus/scribe

CTC loss:

• https://www.dlology.com/blog/how-to-train-a-keras-model-to-recognize-variable-length-text/
• https://hackernoon.com/latest-deep-learning-ocr-with-keras-and-supervisely-in-15-minutes-34aecd630ed8
• https://github.com/SeanNaren/deepspeech.pytorch/blob/master/model.py

Kaggle ctc loss:

• https://dinantdatascientist.blogspot.com/2018/02/kaggle-tensorflow-speech-recognition.html

Explained ctc loss:

• https://gab41.lab41.org/speech-recognition-you-down-with-ctc-8d3b558943f0
• https://distill.pub/2017/ctc/
• https://stats.stackexchange.com/questions/320868/what-is-connectionist-temporal-classification-ctc

CTC loss param:

• https://kur.deepgram.com/specification.html#using-ctc-loss

Overfitting

• https://hackernoon.com/memorizing-is-not-learning-6-tricks-to-prevent-overfitting-in-machine-learning-820b091dc42
• https://elitedatascience.com/overfitting-in-machine-learning