README.md

metadata

library_name: transformers
license: apache-2.0
base_model: facebook/vit-msn-small
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: vit-msn-small-beta-fia-manually-enhanced_test_1
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: default
          split: test
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.7464788732394366

vit-msn-small-beta-fia-manually-enhanced_test_1

This model is a fine-tuned version of facebook/vit-msn-small on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.7014
• Accuracy: 0.7465

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: 1e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 256
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.2
• num_epochs: 50
• label_smoothing_factor: 0.1

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.5714	1	0.9554	0.3732
No log	1.7143	3	0.9304	0.5211
No log	2.8571	5	0.8979	0.6338
No log	4.0	7	0.8770	0.6479
No log	4.5714	8	0.8595	0.6408
0.877	5.7143	10	0.8456	0.5634
0.877	6.8571	12	0.8825	0.4859
0.877	8.0	14	0.8294	0.5563
0.877	8.5714	15	0.7883	0.6197
0.877	9.7143	17	0.7541	0.6549
0.877	10.8571	19	0.7689	0.6690
0.7053	12.0	21	0.7652	0.6620
0.7053	12.5714	22	0.7496	0.6690
0.7053	13.7143	24	0.7139	0.7183
0.7053	14.8571	26	0.7014	0.7465
0.7053	16.0	28	0.7290	0.7183
0.7053	16.5714	29	0.7431	0.6901
0.6176	17.7143	31	0.7498	0.6690
0.6176	18.8571	33	0.7439	0.6761
0.6176	20.0	35	0.7347	0.6972
0.6176	20.5714	36	0.7377	0.6901
0.6176	21.7143	38	0.7227	0.6901
0.6053	22.8571	40	0.7228	0.7042
0.6053	24.0	42	0.7282	0.6901
0.6053	24.5714	43	0.7363	0.6761
0.6053	25.7143	45	0.7431	0.6831
0.6053	26.8571	47	0.7450	0.6831
0.6053	28.0	49	0.7455	0.6761
0.5926	28.5714	50	0.7447	0.6761

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.2.0
• Tokenizers 0.19.1