DunnBC22's picture
Update README.md
27b61d6
metadata
tags:
  - generated_from_trainer
model-index:
  - name: trocr-base-handwritten-OCR-handwriting_recognition_v2
    results: []
language:
  - en
metrics:
  - cer
pipeline_tag: image-to-text

trocr-base-handwritten-OCR-handwriting_recognition_v2

This model is a fine-tuned version of microsoft/trocr-base-handwritten. It achieves the following results on the evaluation set:

  • Loss: 0.2470
  • CER: 0.0360

Model description

For more information on how it was created, check out the following link: https://github.com/DunnBC22/Vision_Audio_and_Multimodal_Projects/blob/main/Optical%20Character%20Recognition%20(OCR)/Handwriting%20Recognition/Handwriting%20Recognition_v2/Mini%20Handwriting%20OCR%20Project.ipynb

Intended uses & limitations

This model is intended to demonstrate my ability to solve a complex problem using technology. You are welcome to test and experiment with this model, but it is at your own risk/peril.

Training and evaluation data

Dataset Source: https://www.kaggle.com/datasets/ssarkar445/handwriting-recognitionocr

Character Length for Training Dataset:

Input Character Length for Training Dataset

Character Length for Evaluation Dataset:

Input Character Length for Evaluation Dataset

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 5e-05
  • train_batch_size: 8
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 3

Training results

Training Loss Epoch Step Validation Loss Cer
0.4292 1.0 2500 0.4332 0.0679
0.2521 2.0 5000 0.2767 0.0483
0.1049 3.0 7500 0.2470 0.0360

Framework versions

  • Transformers 4.26.0
  • Pytorch 1.12.1
  • Datasets 2.8.0
  • Tokenizers 0.12.1