Hugging Face's logo

dennisjooo
/
emotion_classification

Image Classification
Transformers
PyTorch
Safetensors
vit
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Community
9
emotion_classification / README.md
dennisjooo's picture
dennisjooo
End of training
733a9b8
preview code
|
raw
history blame
No virus
6.88 kB
metadata
license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
  - generated_from_trainer
datasets:
  - image_folder
metrics:
  - accuracy
  - precision
  - f1
model-index:
  - name: emotion_classification
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: image_folder
          type: image_folder
          config: FastJobs--Visual_Emotional_Analysis
          split: train
          args: FastJobs--Visual_Emotional_Analysis
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.65625
          - name: Precision
            type: precision
            value: 0.6864745278875714
          - name: F1
            type: f1
            value: 0.6531282051282051

emotion_classification

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the image_folder dataset. It achieves the following results on the evaluation set:

  • Loss: 1.0743
  • Accuracy: 0.6562
  • Precision: 0.6865
  • F1: 0.6531

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: 64
  • eval_batch_size: 64
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine_with_restarts
  • lr_scheduler_warmup_steps: 100
  • num_epochs: 300

Training results

Training Loss Epoch Step Validation Loss Accuracy Precision F1
2.0912 1.0 10 2.0884 0.0938 0.0557 0.0679
2.086 2.0 20 2.0835 0.1062 0.1076 0.0825
2.0724 3.0 30 2.0743 0.15 0.1595 0.1235
2.0575 4.0 40 2.0614 0.1625 0.1451 0.1291
2.0375 5.0 50 2.0399 0.2125 0.2375 0.1880
1.9952 6.0 60 1.9954 0.2875 0.4219 0.2692
1.9309 7.0 70 1.9096 0.3312 0.4116 0.3141
1.8219 8.0 80 1.7690 0.375 0.4091 0.3375
1.6907 9.0 90 1.6323 0.4 0.4548 0.3595
1.5937 10.0 100 1.5317 0.4437 0.4015 0.4174
1.5157 11.0 110 1.4620 0.5312 0.4945 0.5078
1.4458 12.0 120 1.4050 0.5125 0.4734 0.4880
1.3712 13.0 130 1.3719 0.5375 0.5776 0.5236
1.3043 14.0 140 1.3033 0.5687 0.6482 0.5547
1.2424 15.0 150 1.2497 0.5813 0.5970 0.5619
1.2369 16.0 160 1.2423 0.5375 0.4994 0.5061
1.1596 17.0 170 1.2109 0.5563 0.5086 0.5216
1.1252 18.0 180 1.1889 0.5813 0.5772 0.5622
1.0746 19.0 190 1.1752 0.5625 0.5843 0.5631
1.0496 20.0 200 1.1402 0.6062 0.5995 0.5911
0.9874 21.0 210 1.1470 0.5875 0.5897 0.5720
0.9423 22.0 220 1.1294 0.6188 0.6174 0.6072
0.8842 23.0 230 1.1335 0.6 0.6216 0.6004
0.8817 24.0 240 1.1002 0.6 0.6078 0.5970
0.8365 25.0 250 1.1237 0.625 0.6392 0.6209
0.7965 26.0 260 1.1781 0.55 0.5888 0.5419
0.7829 27.0 270 1.1278 0.6 0.6219 0.5947
0.7269 28.0 280 1.1144 0.6 0.6386 0.5937
0.7158 29.0 290 1.1245 0.6125 0.6524 0.5939
0.7178 30.0 300 1.0692 0.6188 0.6344 0.6159
0.6704 31.0 310 1.0568 0.65 0.6724 0.6514
0.6371 32.0 320 1.0411 0.65 0.6529 0.6465
0.6317 33.0 330 1.1018 0.6438 0.6732 0.6416
0.5625 34.0 340 1.0743 0.6562 0.6865 0.6531
0.5717 35.0 350 1.1658 0.6062 0.6636 0.6094
0.5807 36.0 360 1.1473 0.625 0.6654 0.6161
0.5269 37.0 370 1.1367 0.6188 0.6317 0.6150
0.5284 38.0 380 1.0724 0.6438 0.6625 0.6449
0.5715 39.0 390 1.1805 0.575 0.6076 0.5711
0.486 40.0 400 1.1676 0.5938 0.6379 0.5892
0.4581 41.0 410 1.1633 0.6312 0.6583 0.6298
0.4364 42.0 420 1.1371 0.6312 0.6353 0.6255
0.4117 43.0 430 1.2004 0.625 0.6748 0.6086
0.4433 44.0 440 1.1082 0.625 0.6322 0.6232
0.4031 45.0 450 1.2251 0.5875 0.6395 0.5944
0.4205 46.0 460 1.2513 0.5938 0.6196 0.5934
0.3524 47.0 470 1.1704 0.6125 0.6303 0.6147
0.4094 48.0 480 1.1930 0.5875 0.6071 0.5892
0.369 49.0 490 1.1970 0.6188 0.6509 0.6171
0.3666 50.0 500 1.2280 0.6 0.6171 0.5971
0.4054 51.0 510 1.2725 0.5625 0.5807 0.5599
0.4247 52.0 520 1.2380 0.6188 0.6385 0.6128
0.3791 53.0 530 1.2402 0.5813 0.6153 0.5806
0.3241 54.0 540 1.2491 0.5687 0.5817 0.5676
0.3268 55.0 550 1.2575 0.5938 0.6058 0.5956
0.3419 56.0 560 1.3199 0.6 0.6160 0.5930
0.3657 57.0 570 1.2408 0.6188 0.6441 0.6207
0.3327 58.0 580 1.2430 0.6125 0.6200 0.6107
0.3126 59.0 590 1.3995 0.5312 0.5619 0.5158

Framework versions

  • Transformers 4.33.0
  • Pytorch 2.0.0
  • Datasets 2.1.0
  • Tokenizers 0.13.3