RickyIG/emotion_face_image_classification
217
1---2license: apache-2.03base_model: google/vit-base-patch16-224-in21k4tags:5- generated_from_trainer6datasets:7- imagefolder8metrics:9- accuracy10model-index:11- name: emotion_face_image_classification12 results:13 - task:14 name: Image Classification15 type: image-classification16 dataset:17 name: imagefolder18 type: imagefolder19 config: default20 split: train21 args: default22 metrics:23 - name: Accuracy24 type: accuracy25 value: 0.5526---27 28<!-- This model card has been generated automatically according to the information the Trainer had access to. You29should probably proofread and complete it, then remove this comment. -->30 31# emotion_face_image_classification32 33This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the imagefolder dataset.34It achieves the following results on the evaluation set:35- Loss: 1.211036- Accuracy: 0.5537 38## Model description39 40More information needed41 42## Intended uses & limitations43 44More information needed45 46## Training and evaluation data47 48More information needed49 50## Training procedure51 52### Training hyperparameters53 54The following hyperparameters were used during training:55- learning_rate: 5e-0556- train_batch_size: 6457- eval_batch_size: 6458- seed: 4259- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-0860- lr_scheduler_type: linear61- num_epochs: 5062 63### Training results64 65| Training Loss | Epoch | Step | Validation Loss | Accuracy |66|:-------------:|:-----:|:----:|:---------------:|:--------:|67| 2.0717 | 1.0 | 10 | 2.0593 | 0.2062 |68| 2.005 | 2.0 | 20 | 1.9999 | 0.2625 |69| 1.9169 | 3.0 | 30 | 1.8931 | 0.35 |70| 1.7635 | 4.0 | 40 | 1.7616 | 0.4062 |71| 1.6614 | 5.0 | 50 | 1.6452 | 0.4562 |72| 1.6182 | 6.0 | 60 | 1.5661 | 0.4125 |73| 1.5434 | 7.0 | 70 | 1.5183 | 0.4125 |74| 1.46 | 8.0 | 80 | 1.4781 | 0.4875 |75| 1.4564 | 9.0 | 90 | 1.3939 | 0.5125 |76| 1.2966 | 10.0 | 100 | 1.3800 | 0.4562 |77| 1.3732 | 11.0 | 110 | 1.3557 | 0.475 |78| 1.2907 | 12.0 | 120 | 1.3473 | 0.5 |79| 1.2875 | 13.0 | 130 | 1.3416 | 0.5312 |80| 1.2743 | 14.0 | 140 | 1.2964 | 0.4875 |81| 1.1249 | 15.0 | 150 | 1.2385 | 0.525 |82| 1.0963 | 16.0 | 160 | 1.2775 | 0.5062 |83| 1.0261 | 17.0 | 170 | 1.2751 | 0.5125 |84| 0.9298 | 18.0 | 180 | 1.2318 | 0.525 |85| 1.0668 | 19.0 | 190 | 1.2520 | 0.5437 |86| 0.9933 | 20.0 | 200 | 1.2512 | 0.525 |87| 1.1069 | 21.0 | 210 | 1.3016 | 0.5 |88| 1.0279 | 22.0 | 220 | 1.3279 | 0.475 |89| 0.967 | 23.0 | 230 | 1.2481 | 0.5 |90| 0.8115 | 24.0 | 240 | 1.1791 | 0.5563 |91| 0.7912 | 25.0 | 250 | 1.2336 | 0.55 |92| 0.9294 | 26.0 | 260 | 1.1759 | 0.5813 |93| 0.8936 | 27.0 | 270 | 1.1685 | 0.6 |94| 0.7706 | 28.0 | 280 | 1.2403 | 0.5312 |95| 0.7694 | 29.0 | 290 | 1.2479 | 0.5687 |96| 0.7265 | 30.0 | 300 | 1.2000 | 0.5625 |97| 0.6781 | 31.0 | 310 | 1.1856 | 0.55 |98| 0.6676 | 32.0 | 320 | 1.2661 | 0.5437 |99| 0.7254 | 33.0 | 330 | 1.1986 | 0.5437 |100| 0.7396 | 34.0 | 340 | 1.1497 | 0.575 |101| 0.5532 | 35.0 | 350 | 1.2796 | 0.5062 |102| 0.622 | 36.0 | 360 | 1.2749 | 0.5125 |103| 0.6958 | 37.0 | 370 | 1.2034 | 0.5687 |104| 0.6102 | 38.0 | 380 | 1.2576 | 0.5188 |105| 0.6161 | 39.0 | 390 | 1.2635 | 0.5062 |106| 0.6927 | 40.0 | 400 | 1.1535 | 0.5437 |107| 0.549 | 41.0 | 410 | 1.1405 | 0.6 |108| 0.6668 | 42.0 | 420 | 1.2683 | 0.5312 |109| 0.5144 | 43.0 | 430 | 1.2249 | 0.6 |110| 0.6703 | 44.0 | 440 | 1.2297 | 0.5687 |111| 0.6383 | 45.0 | 450 | 1.1507 | 0.6062 |112| 0.5211 | 46.0 | 460 | 1.2914 | 0.4813 |113| 0.4743 | 47.0 | 470 | 1.2782 | 0.5125 |114| 0.553 | 48.0 | 480 | 1.2256 | 0.5375 |115| 0.6407 | 49.0 | 490 | 1.2149 | 0.5687 |116| 0.4195 | 50.0 | 500 | 1.2024 | 0.5625 |117 118 119### Framework versions120 121- Transformers 4.33.2122- Pytorch 2.0.1+cu118123- Datasets 2.14.5124- Tokenizers 0.13.3125 