square_run_32_batch

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

  • Loss: 1.6241
  • F1 Macro: 0.5019
  • F1 Micro: 0.5758
  • F1 Weighted: 0.5679
  • Precision Macro: 0.5021
  • Precision Micro: 0.5758
  • Precision Weighted: 0.5657
  • Recall Macro: 0.5073
  • Recall Micro: 0.5758
  • Recall Weighted: 0.5758
  • Accuracy: 0.5758

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: 0.0001
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 42
  • optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 30

Training results

Training Loss Epoch Step Validation Loss F1 Macro F1 Micro F1 Weighted Precision Macro Precision Micro Precision Weighted Recall Macro Recall Micro Recall Weighted Accuracy
1.9373 1.0 15 1.8818 0.0464 0.1894 0.0615 0.0277 0.1894 0.0367 0.1429 0.1894 0.1894 0.1894
1.869 2.0 30 1.8642 0.1100 0.2652 0.1418 0.075 0.2652 0.0968 0.2063 0.2652 0.2652 0.2652
1.9218 3.0 45 1.8754 0.1163 0.2576 0.1460 0.1316 0.2576 0.1566 0.1905 0.2576 0.2576 0.2576
1.6733 4.0 60 1.6881 0.2445 0.3864 0.3053 0.2427 0.3864 0.2917 0.2992 0.3864 0.3864 0.3864
1.54 5.0 75 1.5528 0.3252 0.4242 0.3856 0.3429 0.4242 0.4101 0.3570 0.4242 0.4242 0.4242
1.4418 6.0 90 1.5737 0.2858 0.3864 0.3213 0.2846 0.3864 0.3243 0.3398 0.3864 0.3864 0.3864
0.8592 7.0 105 1.5408 0.3444 0.4394 0.3965 0.3208 0.4394 0.3674 0.3791 0.4394 0.4394 0.4394
1.1427 8.0 120 1.2804 0.4638 0.5606 0.5317 0.4698 0.5606 0.5280 0.4831 0.5606 0.5606 0.5606
0.7849 9.0 135 1.2880 0.4649 0.5530 0.5291 0.4804 0.5530 0.5401 0.4823 0.5530 0.5530 0.5530
0.6846 10.0 150 1.3130 0.4298 0.5152 0.4811 0.4404 0.5152 0.5005 0.4671 0.5152 0.5152 0.5152
0.4006 11.0 165 1.2958 0.4931 0.5833 0.5598 0.4983 0.5833 0.5756 0.5229 0.5833 0.5833 0.5833
0.4329 12.0 180 1.2990 0.5062 0.5530 0.5562 0.5315 0.5530 0.5874 0.5133 0.5530 0.5530 0.5530
0.482 13.0 195 1.3831 0.4842 0.5152 0.5233 0.5517 0.5152 0.5803 0.4839 0.5152 0.5152 0.5152
0.6409 14.0 210 1.4066 0.5081 0.5985 0.5765 0.5194 0.5985 0.5820 0.5232 0.5985 0.5985 0.5985
0.3206 15.0 225 1.3690 0.5155 0.5606 0.5520 0.6158 0.5606 0.5890 0.5170 0.5606 0.5606 0.5606
0.1773 16.0 240 1.2568 0.5920 0.6515 0.6408 0.6894 0.6515 0.6623 0.5843 0.6515 0.6515 0.6515
0.3259 17.0 255 1.3406 0.5467 0.6061 0.5961 0.5615 0.6061 0.6033 0.5467 0.6061 0.6061 0.6061
0.1123 18.0 270 1.3767 0.5868 0.6364 0.6306 0.6258 0.6364 0.6413 0.5785 0.6364 0.6364 0.6364
0.1129 19.0 285 1.4680 0.5879 0.6439 0.6306 0.6809 0.6439 0.6933 0.5806 0.6439 0.6439 0.6439
0.0651 20.0 300 1.4981 0.6655 0.6894 0.6876 0.7115 0.6894 0.7224 0.6511 0.6894 0.6894 0.6894
0.0685 21.0 315 1.4621 0.6091 0.6515 0.6494 0.6303 0.6515 0.6641 0.6040 0.6515 0.6515 0.6515
0.1469 22.0 330 1.5347 0.5330 0.6212 0.6040 0.5477 0.6212 0.6149 0.5440 0.6212 0.6212 0.6212
0.0289 23.0 345 1.5417 0.5466 0.6288 0.6180 0.5409 0.6288 0.6108 0.5549 0.6288 0.6288 0.6288
0.01 24.0 360 1.5670 0.5475 0.6364 0.6187 0.5435 0.6364 0.6104 0.5594 0.6364 0.6364 0.6364
0.035 25.0 375 1.6037 0.5529 0.6364 0.6209 0.5470 0.6364 0.6156 0.5679 0.6364 0.6364 0.6364
0.0109 26.0 390 1.6752 0.5897 0.6212 0.6203 0.6145 0.6212 0.6527 0.6000 0.6212 0.6212 0.6212
0.038 27.0 405 1.6724 0.5344 0.6136 0.6008 0.5332 0.6136 0.6005 0.5468 0.6136 0.6136 0.6136
0.0116 28.0 420 1.6252 0.5384 0.6212 0.6090 0.5337 0.6212 0.6033 0.5491 0.6212 0.6212 0.6212
0.006 29.0 435 1.5980 0.5572 0.6364 0.6294 0.5529 0.6364 0.6246 0.5634 0.6364 0.6364 0.6364
0.0046 30.0 450 1.5939 0.5605 0.6439 0.6342 0.5546 0.6439 0.6269 0.5687 0.6439 0.6439 0.6439

Framework versions

  • Transformers 4.48.2
  • Pytorch 2.6.0+cu124
  • Datasets 3.2.0
  • Tokenizers 0.21.0
Downloads last month
13
Safetensors
Model size
85.8M params
Tensor type
F32
·
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.

Model tree for corranm/square_run_32_batch

Finetuned
(633)
this model