metadata

license: apache-2.0
base_model: microsoft/swinv2-base-patch4-window16-256
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - precision
  - recall
  - f1
model-index:
  - name: swinv2-base-patch4-window16-256-finetuned-galaxy10-decals
    results: []

swinv2-base-patch4-window16-256-finetuned-galaxy10-decals

This model is a fine-tuned version of microsoft/swinv2-base-patch4-window16-256 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.4826
Accuracy: 0.8557
Precision: 0.8544
Recall: 0.8557
F1: 0.8543

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
gradient_accumulation_steps: 4
total_train_batch_size: 256
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
1.5098	0.99	62	1.2358	0.5569	0.5493	0.5569	0.5321
0.8845	2.0	125	0.7391	0.7599	0.7800	0.7599	0.7497
0.753	2.99	187	0.5997	0.7971	0.8062	0.7971	0.7903
0.6149	4.0	250	0.4920	0.8331	0.8285	0.8331	0.8276
0.5807	4.99	312	0.4623	0.8326	0.8323	0.8326	0.8315
0.5938	6.0	375	0.4857	0.8365	0.8403	0.8365	0.8294
0.5583	6.99	437	0.4680	0.8264	0.8314	0.8264	0.8243
0.5103	8.0	500	0.4882	0.8191	0.8312	0.8191	0.8180
0.5186	8.99	562	0.4341	0.8574	0.8589	0.8574	0.8546
0.4696	10.0	625	0.4293	0.8495	0.8484	0.8495	0.8481
0.4711	10.99	687	0.4396	0.8422	0.8431	0.8422	0.8414
0.4271	12.0	750	0.4547	0.8489	0.8500	0.8489	0.8480
0.4576	12.99	812	0.4424	0.8489	0.8522	0.8489	0.8473
0.4483	14.0	875	0.4355	0.8495	0.8531	0.8495	0.8492
0.3914	14.99	937	0.4360	0.8540	0.8533	0.8540	0.8532
0.3883	16.0	1000	0.4464	0.8546	0.8550	0.8546	0.8526
0.3421	16.99	1062	0.4473	0.8489	0.8486	0.8489	0.8479
0.3666	18.0	1125	0.4455	0.8540	0.8541	0.8540	0.8528
0.3737	18.99	1187	0.4587	0.8574	0.8560	0.8574	0.8561
0.3694	20.0	1250	0.4583	0.8551	0.8528	0.8551	0.8523
0.3269	20.99	1312	0.4883	0.8506	0.8494	0.8506	0.8487
0.3699	22.0	1375	0.4808	0.8501	0.8514	0.8501	0.8486
0.3395	22.99	1437	0.4706	0.8484	0.8493	0.8484	0.8477
0.3147	24.0	1500	0.4676	0.8568	0.8556	0.8568	0.8557
0.3352	24.99	1562	0.4868	0.8557	0.8543	0.8557	0.8538
0.3007	26.0	1625	0.4887	0.8489	0.8492	0.8489	0.8475
0.3049	26.99	1687	0.4838	0.8534	0.8532	0.8534	0.8526
0.3228	28.0	1750	0.4910	0.8551	0.8539	0.8551	0.8536
0.3005	28.99	1812	0.4846	0.8534	0.8517	0.8534	0.8518
0.2972	29.76	1860	0.4826	0.8557	0.8544	0.8557	0.8543

Framework versions

Transformers 4.37.2
Pytorch 2.3.0
Datasets 2.19.1
Tokenizers 0.15.1