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
f717422
preview code
|
raw
history blame
6.82 kB
metadata
license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
  - precision
  - f1
model-index:
  - name: emotion_classification
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: default
          split: train
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.64375
          - name: Precision
            type: precision
            value: 0.650616883116883
          - name: F1
            type: f1
            value: 0.6344950707077283

emotion_classification

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

  • Loss: 1.1553
  • Accuracy: 0.6438
  • Precision: 0.6506
  • F1: 0.6345

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: 3e-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: 150
  • num_epochs: 300

Training results

Training Loss Epoch Step Validation Loss Accuracy Precision F1
2.0799 1.0 10 2.0707 0.1313 0.1740 0.1156
2.0811 2.0 20 2.0681 0.1437 0.1617 0.1245
2.0709 3.0 30 2.0640 0.1562 0.1544 0.1330
2.0701 4.0 40 2.0590 0.1688 0.1463 0.1431
2.0639 5.0 50 2.0529 0.1812 0.1676 0.1613
2.0499 6.0 60 2.0439 0.2 0.2050 0.1871
2.0387 7.0 70 2.0322 0.25 0.2679 0.2373
2.0235 8.0 80 2.0141 0.3312 0.3638 0.3331
1.9933 9.0 90 1.9883 0.3375 0.3752 0.3392
1.9573 10.0 100 1.9473 0.3563 0.3940 0.3535
1.912 11.0 110 1.8863 0.3875 0.4352 0.3759
1.8306 12.0 120 1.8102 0.3875 0.4062 0.3586
1.7479 13.0 130 1.7158 0.4062 0.4056 0.3689
1.665 14.0 140 1.6250 0.475 0.4543 0.4248
1.6115 15.0 150 1.5597 0.4875 0.4646 0.4414
1.5716 16.0 160 1.5112 0.5125 0.4846 0.4575
1.5062 17.0 170 1.4672 0.525 0.4932 0.4925
1.4655 18.0 180 1.4262 0.5312 0.5018 0.4876
1.413 19.0 190 1.3851 0.575 0.5253 0.5317
1.3758 20.0 200 1.3421 0.5625 0.5900 0.5113
1.317 21.0 210 1.3156 0.55 0.5835 0.4996
1.291 22.0 220 1.2712 0.5938 0.6374 0.5601
1.2369 23.0 230 1.2697 0.5563 0.5681 0.5250
1.2139 24.0 240 1.2439 0.5625 0.5733 0.5417
1.1766 25.0 250 1.2228 0.5938 0.6099 0.5735
1.1483 26.0 260 1.2464 0.5625 0.6016 0.5508
1.1344 27.0 270 1.1877 0.5875 0.6142 0.5718
1.0898 28.0 280 1.1871 0.6 0.6481 0.5817
1.0515 29.0 290 1.1553 0.6438 0.6506 0.6345
1.0628 30.0 300 1.1603 0.575 0.6209 0.5727
1.0257 31.0 310 1.1326 0.6125 0.6312 0.6109
1.0048 32.0 320 1.1450 0.6125 0.6402 0.6079
0.9646 33.0 330 1.1250 0.6062 0.6161 0.6004
0.9231 34.0 340 1.1299 0.6 0.6183 0.5976
0.8944 35.0 350 1.1312 0.5938 0.5996 0.5885
0.9001 36.0 360 1.1293 0.625 0.6358 0.6220
0.8587 37.0 370 1.1415 0.6062 0.6122 0.6037
0.8708 38.0 380 1.1171 0.6062 0.6379 0.5985
0.843 39.0 390 1.1220 0.625 0.6658 0.6229
0.8038 40.0 400 1.1144 0.6188 0.6243 0.6153
0.7815 41.0 410 1.1538 0.575 0.6042 0.5679
0.7289 42.0 420 1.1125 0.6062 0.6218 0.6024
0.7255 43.0 430 1.1401 0.6 0.6307 0.5947
0.7182 44.0 440 1.1092 0.6 0.6121 0.5916
0.6533 45.0 450 1.1219 0.625 0.6448 0.6268
0.6658 46.0 460 1.1322 0.6125 0.6272 0.6135
0.6293 47.0 470 1.1306 0.6 0.6075 0.5980
0.6287 48.0 480 1.1227 0.6125 0.6210 0.6099
0.622 49.0 490 1.1441 0.5938 0.6154 0.5940
0.6004 50.0 500 1.1119 0.625 0.6267 0.6206
0.606 51.0 510 1.1301 0.5938 0.6146 0.5925
0.5924 52.0 520 1.1552 0.6062 0.6135 0.6022
0.5639 53.0 530 1.1956 0.5938 0.6411 0.5945
0.5434 54.0 540 1.1843 0.5813 0.5925 0.5765
0.5479 55.0 550 1.1529 0.6125 0.6247 0.6142
0.5227 56.0 560 1.1730 0.5687 0.5724 0.5628
0.5402 57.0 570 1.1919 0.6 0.6075 0.5954
0.4971 58.0 580 1.1761 0.5938 0.5984 0.5925
0.5004 59.0 590 1.2305 0.5687 0.5957 0.5645

Framework versions

  • Transformers 4.33.1
  • Pytorch 2.0.0
  • Datasets 2.14.5
  • Tokenizers 0.13.3