metadata

license: apache-2.0
base_model: microsoft/swinv2-base-patch4-window16-256
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
  - precision
  - recall
model-index:
  - name: SwinV2-Base-Document-Classifier
    results: []

SwinV2-Base-Document-Classifier

This model is a fine-tuned version of microsoft/swinv2-base-patch4-window16-256 on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.0573
Accuracy: 0.9810
F1: 0.9810
Precision: 0.9813
Recall: 0.9810

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
training_steps: 400

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
0.0814	0.12	50	0.0876	0.9740	0.9741	0.9749	0.9740
0.078	0.13	51	0.0658	0.9797	0.9797	0.9799	0.9797
0.143	0.13	52	0.0609	0.9831	0.9831	0.9832	0.9831
0.0863	0.13	53	0.0595	0.9832	0.9832	0.9834	0.9832
0.0561	0.14	54	0.0650	0.9825	0.9825	0.9826	0.9825
0.1118	0.14	55	0.0697	0.9804	0.9804	0.9806	0.9804
0.1258	0.14	56	0.1053	0.9678	0.9678	0.9697	0.9678
0.132	0.14	57	0.0672	0.9781	0.9781	0.9785	0.9781
0.1564	0.14	58	0.0582	0.9827	0.9827	0.9829	0.9827
0.1085	0.15	59	0.0557	0.9849	0.9848	0.9849	0.9848
0.0506	0.15	60	0.0548	0.9845	0.9845	0.9845	0.9845
0.0787	0.15	61	0.1049	0.9660	0.9661	0.9681	0.9660
0.1211	0.15	62	0.0792	0.9741	0.9741	0.9746	0.9741
0.0907	0.16	63	0.0608	0.9826	0.9825	0.9826	0.9826
0.0419	0.16	64	0.0577	0.9832	0.9832	0.9833	0.9832
0.1659	0.16	65	0.0548	0.9843	0.9843	0.9844	0.9843
0.0983	0.17	66	0.0657	0.9805	0.9805	0.9814	0.9804
0.0535	0.17	67	0.0493	0.9851	0.9851	0.9851	0.9851
0.0936	0.17	68	0.0588	0.9827	0.9827	0.9830	0.9827
0.0933	0.17	69	0.0573	0.9810	0.9810	0.9813	0.9810

Framework versions

Transformers 4.39.3
Pytorch 2.2.2
Datasets 2.18.0
Tokenizers 0.15.2