Hugging Face's logo

egujr001
/
egujr001-swim2-base-model

Image Classification
Transformers
PyTorch
swinv2
Generated from Trainer
Inference Endpoints
Model card Files Community
egujr001-swim2-base-model / README.md
egujr001's picture
egujr001
egujr001/egujr001-AquaAnalyser-model
2a61a2f verified
preview code
|
raw history blame contribute delete
No virus
6.19 kB
metadata
license: apache-2.0
base_model: microsoft/swinv2-tiny-patch4-window8-256
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: egujr001-swim2-base-model
    results: []

egujr001-swim2-base-model

This model is a fine-tuned version of microsoft/swinv2-tiny-patch4-window8-256 on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1969
  • Accuracy: 0.9457

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: 56
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 3

Training results

Training Loss Epoch Step Validation Loss Accuracy
0.6254 0.04 100 0.5840 0.7050
0.3998 0.08 200 0.3525 0.8507
0.2796 0.12 300 0.2710 0.8975
0.23 0.15 400 0.2660 0.9012
0.2372 0.19 500 0.1678 0.9401
0.1944 0.23 600 0.1437 0.9437
0.1635 0.27 700 0.1231 0.9503
0.1463 0.31 800 0.1353 0.9551
0.1287 0.35 900 0.1216 0.9523
0.1208 0.39 1000 0.1695 0.9351
0.1204 0.42 1100 0.1221 0.9557
0.1064 0.46 1200 0.1605 0.9432
0.1114 0.5 1300 0.0998 0.9613
0.1324 0.54 1400 0.0888 0.9650
0.0997 0.58 1500 0.0810 0.9686
0.0904 0.62 1600 0.0945 0.9655
0.0975 0.66 1700 0.0978 0.9635
0.0859 0.69 1800 0.0858 0.9696
0.0785 0.73 1900 0.0749 0.9722
0.0743 0.77 2000 0.0763 0.9727
0.0815 0.81 2100 0.0765 0.9728
0.0674 0.85 2200 0.0881 0.9703
0.0726 0.89 2300 0.0875 0.9716
0.0633 0.93 2400 0.0912 0.9721
0.0501 0.96 2500 0.0743 0.9750
0.0927 1.0 2600 0.0695 0.9759
0.0766 1.04 2700 0.0788 0.9733
0.0934 1.08 2800 0.0699 0.9753
0.0714 1.12 2900 0.0756 0.9762
0.069 1.16 3000 0.0859 0.9706
0.0702 1.2 3100 0.1001 0.9658
0.0633 1.23 3200 0.0724 0.9756
0.0756 1.27 3300 0.0734 0.9745
0.0617 1.31 3400 0.0704 0.9747
0.0498 1.35 3500 0.0651 0.9788
0.0668 1.39 3600 0.0625 0.9791
0.0441 1.43 3700 0.0714 0.9774
0.0789 1.46 3800 0.0880 0.9722
0.0464 1.5 3900 0.0720 0.9749
0.0532 1.54 4000 0.0681 0.9782
0.0677 1.58 4100 0.0733 0.9736
0.0654 1.62 4200 0.0610 0.9802
0.0554 1.66 4300 0.0825 0.9740
0.0836 1.7 4400 0.0694 0.9780
0.0688 1.73 4500 0.0599 0.9813
0.052 1.77 4600 0.0932 0.9673
0.0515 1.81 4700 0.0785 0.9759
0.0586 1.85 4800 0.0660 0.9787
0.056 1.89 4900 0.0612 0.9783
0.037 1.93 5000 0.0645 0.9795
0.0541 1.97 5100 0.0600 0.9809
0.0521 2.0 5200 0.0876 0.9737
0.0352 2.04 5300 0.0709 0.9780
0.0498 2.08 5400 0.0610 0.9809
0.0424 2.12 5500 0.0569 0.9830
0.0532 2.16 5600 0.0625 0.9820
0.046 2.2 5700 0.0512 0.9842
0.0453 2.24 5800 0.0608 0.9813
0.0577 2.27 5900 0.0697 0.9811
0.0397 2.31 6000 0.0688 0.9816
0.0494 2.35 6100 0.0534 0.9834
0.0158 2.39 6200 0.0860 0.9774
0.0297 2.43 6300 0.0593 0.9836
0.055 2.47 6400 0.0579 0.9821
0.0368 2.51 6500 0.0729 0.9796
0.0754 2.54 6600 0.0601 0.9827
0.0523 2.58 6700 0.0597 0.9824
0.0433 2.62 6800 0.0547 0.9841
0.0164 2.66 6900 0.0620 0.9827
0.015 2.7 7000 0.0639 0.9822
0.0415 2.74 7100 0.0576 0.9837
0.0257 2.78 7200 0.0620 0.9820
0.0268 2.81 7300 0.0568 0.9837
0.043 2.85 7400 0.0558 0.9836
0.0339 2.89 7500 0.0554 0.9839
0.0263 2.93 7600 0.0552 0.9837
0.0428 2.97 7700 0.0535 0.9842

Framework versions

  • Transformers 4.33.0
  • Pytorch 2.0.0+cu117
  • Datasets 2.19.1
  • Tokenizers 0.13.3