swinv2-tiny-patch4-window8-256-dmae-humeda-DAV5

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:

Model description

More information needed

More information needed

More information needed

The following hyperparameters were used during training:

learning_rate: 3e-05
train_batch_size: 32
eval_batch_size: 32
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 64
optimizer: Use OptimizerNames.ADAMW_TORCH 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: 0.05
num_epochs: 40

Training Loss	Epoch	Step	Validation Loss	Accuracy
3.0169	1.0	10	1.4927	0.3654
2.7431	2.0	20	1.2785	0.5577
2.1791	3.0	30	1.0628	0.5769
1.7293	4.0	40	1.0541	0.5577
1.3855	5.0	50	0.9817	0.6346
1.3507	6.0	60	0.9506	0.6154
1.186	7.0	70	0.9259	0.5962
1.1201	8.0	80	0.8867	0.6346
1.0185	9.0	90	0.9556	0.5769
0.981	10.0	100	0.9069	0.6923
0.8472	11.0	110	0.9413	0.6346
0.7333	12.0	120	0.9594	0.6346
0.7	13.0	130	0.9818	0.6538
0.7062	14.0	140	0.9588	0.6154
0.6666	15.0	150	1.1824	0.5769
0.6756	16.0	160	1.0542	0.6538
0.6148	17.0	170	1.0112	0.6346
0.6123	18.0	180	1.2390	0.5769
0.5931	19.0	190	1.0358	0.6923
0.5242	20.0	200	1.1471	0.6346
0.5499	21.0	210	1.0452	0.6731
0.4806	22.0	220	1.0887	0.6346
0.4294	23.0	230	1.1078	0.6346
0.5176	24.0	240	1.1218	0.5769
0.4051	25.0	250	1.1255	0.6731
0.4486	26.0	260	1.0775	0.6346
0.4262	27.0	270	1.0711	0.6731
0.4717	28.0	280	1.0975	0.6346
0.4067	29.0	290	1.0647	0.6731
0.3691	30.0	300	1.1139	0.6346
0.4446	31.0	310	1.1270	0.5962
0.3399	32.0	320	1.1498	0.6346
0.3449	33.0	330	1.1864	0.6346
0.4118	34.0	340	1.1989	0.5962
0.3945	35.0	350	1.1928	0.5962
0.3609	36.0	360	1.1815	0.6346

Safetensors

Model size

27.6M params

Tensor type

F32

Base model

Finetuned

(138)

this model