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
116ea51
preview code
|
raw
history blame
No virus
7.3 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: 0.66875
          - name: Precision
            type: precision
            value: 0.684222027972028
          - name: F1
            type: f1
            value: 0.6649370603045093

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: 1.0254
  • Accuracy: 0.6687
  • Precision: 0.6842
  • F1: 0.6649

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: 150
  • num_epochs: 300

Training results

Training Loss Epoch Step Validation Loss Accuracy Precision F1
2.079 1.0 10 2.0759 0.1437 0.1305 0.1297
2.0798 2.0 20 2.0725 0.1688 0.1495 0.1503
2.0758 3.0 30 2.0668 0.2 0.1992 0.1859
2.0602 4.0 40 2.0595 0.225 0.2219 0.2100
2.0456 5.0 50 2.0495 0.225 0.2285 0.2105
2.0303 6.0 60 2.0324 0.2437 0.2546 0.2267
2.0009 7.0 70 1.9983 0.2437 0.2661 0.2291
1.9482 8.0 80 1.9342 0.3375 0.3320 0.3183
1.8709 9.0 90 1.8475 0.4 0.3524 0.3583
1.7828 10.0 100 1.7259 0.4562 0.4074 0.4111
1.6841 11.0 110 1.6324 0.4688 0.4211 0.4182
1.6047 12.0 120 1.5508 0.4375 0.4049 0.3908
1.5343 13.0 130 1.4942 0.5188 0.5115 0.4980
1.4606 14.0 140 1.4133 0.55 0.5063 0.5083
1.3935 15.0 150 1.3513 0.5312 0.5377 0.5050
1.3695 16.0 160 1.2981 0.6062 0.6190 0.5899
1.2956 17.0 170 1.2630 0.5687 0.5654 0.5479
1.2481 18.0 180 1.2470 0.5875 0.5931 0.5735
1.2084 19.0 190 1.2095 0.5938 0.6143 0.5899
1.1676 20.0 200 1.1918 0.5938 0.6006 0.5788
1.0999 21.0 210 1.2066 0.5875 0.6020 0.5690
1.071 22.0 220 1.1474 0.6 0.5997 0.5852
0.9925 23.0 230 1.1266 0.6312 0.6504 0.6283
0.961 24.0 240 1.1031 0.5938 0.6021 0.5901
0.9364 25.0 250 1.1458 0.6 0.6199 0.5907
0.8906 26.0 260 1.1339 0.5875 0.6158 0.5789
0.882 27.0 270 1.0824 0.6312 0.6543 0.6303
0.827 28.0 280 1.1464 0.5875 0.6521 0.5793
0.7791 29.0 290 1.1309 0.575 0.5998 0.5566
0.7621 30.0 300 1.0579 0.6125 0.6277 0.6068
0.7245 31.0 310 1.0418 0.6562 0.6633 0.6533
0.6868 32.0 320 1.0555 0.6375 0.6470 0.6329
0.653 33.0 330 1.1451 0.5938 0.6330 0.5944
0.6102 34.0 340 1.0254 0.6687 0.6842 0.6649
0.5977 35.0 350 1.0981 0.625 0.6482 0.6227
0.6258 36.0 360 1.0975 0.6438 0.6773 0.6346
0.5444 37.0 370 1.1195 0.6125 0.6408 0.6147
0.5558 38.0 380 1.0637 0.625 0.6323 0.6201
0.5716 39.0 390 1.1407 0.6062 0.6463 0.6111
0.5048 40.0 400 1.1153 0.6312 0.6407 0.6244
0.4646 41.0 410 1.1072 0.625 0.6284 0.6225
0.463 42.0 420 1.1086 0.6062 0.6062 0.6026
0.4321 43.0 430 1.1725 0.6 0.6304 0.5960
0.49 44.0 440 1.1325 0.6188 0.6423 0.6166
0.408 45.0 450 1.2134 0.575 0.5865 0.5721
0.4296 46.0 460 1.2182 0.6188 0.6492 0.6175
0.3328 47.0 470 1.1789 0.6188 0.6378 0.6205
0.3781 48.0 480 1.2054 0.6125 0.6158 0.6077
0.3326 49.0 490 1.2308 0.5938 0.6148 0.5941
0.3526 50.0 500 1.2640 0.6 0.6038 0.5959
0.3967 51.0 510 1.3154 0.5437 0.5635 0.5410
0.4286 52.0 520 1.2358 0.6188 0.6488 0.6140
0.3411 53.0 530 1.1959 0.625 0.6368 0.6192
0.3455 54.0 540 1.2526 0.6 0.6168 0.5973
0.3224 55.0 550 1.1988 0.625 0.6490 0.6208
0.3015 56.0 560 1.2067 0.6062 0.6030 0.6005
0.322 57.0 570 1.2124 0.6188 0.6279 0.6181
0.2991 58.0 580 1.2274 0.6312 0.6368 0.6294
0.3199 59.0 590 1.2649 0.5938 0.5876 0.5880
0.3204 60.0 600 1.2636 0.6062 0.6239 0.6002
0.2831 61.0 610 1.3039 0.5875 0.5974 0.5832
0.2723 62.0 620 1.2620 0.625 0.6558 0.6236
0.2806 63.0 630 1.2368 0.6312 0.6364 0.6294
0.2621 64.0 640 1.2783 0.6062 0.6160 0.6049

Framework versions

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