ditmodel / README.md
SonishMaharjan's picture
Update README.md
939b6fe verified
metadata
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: ditmodel
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: test
          split: train
          args: test
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.9523326572008114

ditmodel

This model was fintuned on DiT model for document classification on custom dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1482
  • Accuracy: 0.9523
  • Weighted f1: 0.9524
  • Micro f1: 0.9523
  • Macro f1: 0.9505
  • Weighted recall: 0.9523
  • Micro recall: 0.9523
  • Macro recall: 0.9523
  • Weighted precision: 0.9544
  • Micro precision: 0.9523
  • Macro precision: 0.9506

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 5e-05
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 128
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 3

Training results

Training Loss Epoch Step Validation Loss Accuracy Weighted f1 Micro f1 Macro f1 Weighted recall Micro recall Macro recall Weighted precision Micro precision Macro precision
0.2337 1.0 78 0.2668 0.9087 0.9098 0.9087 0.9058 0.9087 0.9087 0.9040 0.9229 0.9087 0.9220
0.1711 2.0 156 0.1820 0.9376 0.9380 0.9376 0.9331 0.9376 0.9376 0.9403 0.9416 0.9376 0.9292
0.1297 3.0 234 0.1482 0.9523 0.9524 0.9523 0.9505 0.9523 0.9523 0.9523 0.9544 0.9523 0.9506

Framework versions

  • Transformers 4.35.2
  • Pytorch 2.1.0+cu121
  • Datasets 2.6.1
  • Tokenizers 0.15.1