metadata
license: apache-2.0
base_model: microsoft/beit-base-patch16-224-pt22k-ft22k
tags:
- generated_from_trainer
datasets:
- imagefolder
metrics:
- accuracy
model-index:
- name: finetuned-FER2013
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: 0.7133402995471961
finetuned-FER2013
This model is a fine-tuned version of microsoft/beit-base-patch16-224-pt22k-ft22k on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.8079
- Accuracy: 0.7133
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-06
- 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
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 60
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.7999 | 1.0 | 202 | 1.7896 | 0.2713 |
| 1.5165 | 2.0 | 404 | 1.4311 | 0.4758 |
| 1.3381 | 3.0 | 606 | 1.2065 | 0.5468 |
| 1.2629 | 4.0 | 808 | 1.1050 | 0.5886 |
| 1.248 | 5.0 | 1010 | 1.0425 | 0.6054 |
| 1.2007 | 6.0 | 1212 | 0.9874 | 0.6339 |
| 1.083 | 7.0 | 1414 | 0.9610 | 0.6447 |
| 1.1061 | 8.0 | 1616 | 0.9385 | 0.6524 |
| 1.0597 | 9.0 | 1818 | 0.9155 | 0.6517 |
| 1.0511 | 10.0 | 2020 | 0.9128 | 0.6580 |
| 1.012 | 11.0 | 2222 | 0.9048 | 0.6660 |
| 1.0479 | 12.0 | 2424 | 0.8821 | 0.6729 |
| 0.9993 | 13.0 | 2626 | 0.8770 | 0.6747 |
| 0.9784 | 14.0 | 2828 | 0.8672 | 0.6757 |
| 1.0439 | 15.0 | 3030 | 0.8766 | 0.6750 |
| 0.9782 | 16.0 | 3232 | 0.8658 | 0.6747 |
| 0.9664 | 17.0 | 3434 | 0.8596 | 0.6764 |
| 1.0132 | 18.0 | 3636 | 0.8491 | 0.6806 |
| 0.9703 | 19.0 | 3838 | 0.8538 | 0.6827 |
| 0.9399 | 20.0 | 4040 | 0.8452 | 0.6876 |
| 0.9299 | 21.0 | 4242 | 0.8420 | 0.6904 |
| 0.9815 | 22.0 | 4444 | 0.8417 | 0.6872 |
| 0.9029 | 23.0 | 4646 | 0.8379 | 0.6900 |
| 0.9142 | 24.0 | 4848 | 0.8336 | 0.6897 |
| 0.8695 | 25.0 | 5050 | 0.8312 | 0.6938 |
| 0.8791 | 26.0 | 5252 | 0.8323 | 0.6942 |
| 0.923 | 27.0 | 5454 | 0.8244 | 0.6956 |
| 0.8866 | 28.0 | 5656 | 0.8261 | 0.6970 |
| 0.9319 | 29.0 | 5858 | 0.8255 | 0.6991 |
| 0.9019 | 30.0 | 6060 | 0.8160 | 0.7050 |
| 0.8785 | 31.0 | 6262 | 0.8169 | 0.7071 |
| 0.8859 | 32.0 | 6464 | 0.8178 | 0.7039 |
| 0.8464 | 33.0 | 6666 | 0.8147 | 0.7092 |
| 0.9143 | 34.0 | 6868 | 0.8232 | 0.7029 |
| 0.8506 | 35.0 | 7070 | 0.8158 | 0.7032 |
| 0.9084 | 36.0 | 7272 | 0.8166 | 0.7057 |
| 0.8616 | 37.0 | 7474 | 0.8132 | 0.7088 |
| 0.8656 | 38.0 | 7676 | 0.8155 | 0.7046 |
| 0.8238 | 39.0 | 7878 | 0.8170 | 0.7064 |
| 0.8673 | 40.0 | 8080 | 0.8190 | 0.7092 |
| 0.8624 | 41.0 | 8282 | 0.8127 | 0.7095 |
| 0.8261 | 42.0 | 8484 | 0.8113 | 0.7113 |
| 0.8218 | 43.0 | 8686 | 0.8150 | 0.7095 |
| 0.8584 | 44.0 | 8888 | 0.8170 | 0.7071 |
| 0.8156 | 45.0 | 9090 | 0.8117 | 0.7119 |
| 0.8075 | 46.0 | 9292 | 0.8133 | 0.7116 |
| 0.8382 | 47.0 | 9494 | 0.8146 | 0.7088 |
| 0.7501 | 48.0 | 9696 | 0.8096 | 0.7113 |
| 0.7859 | 49.0 | 9898 | 0.8102 | 0.7081 |
| 0.8195 | 50.0 | 10100 | 0.8121 | 0.7085 |
| 0.8397 | 51.0 | 10302 | 0.8120 | 0.7099 |
| 0.8561 | 52.0 | 10504 | 0.8089 | 0.7126 |
| 0.8082 | 53.0 | 10706 | 0.8090 | 0.7133 |
| 0.8574 | 54.0 | 10908 | 0.8087 | 0.7106 |
| 0.8611 | 55.0 | 11110 | 0.8093 | 0.7092 |
| 0.8886 | 56.0 | 11312 | 0.8100 | 0.7092 |
| 0.7857 | 57.0 | 11514 | 0.8086 | 0.7133 |
| 0.8467 | 58.0 | 11716 | 0.8083 | 0.7119 |
| 0.795 | 59.0 | 11918 | 0.8083 | 0.7119 |
| 0.7975 | 60.0 | 12120 | 0.8079 | 0.7133 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.0
- Tokenizers 0.15.0