metadata

license: apache-2.0
base_model: microsoft/swin-base-patch4-window7-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: swin-base-patch4-window7-224-finetuned-st-wsdmhar
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: default
          split: train
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.7937443336355394

swin-base-patch4-window7-224-finetuned-st-wsdmhar

This model is a fine-tuned version of microsoft/swin-base-patch4-window7-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

Loss: 0.8737
Accuracy: 0.7937

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: 32
eval_batch_size: 32
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 128
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.7172	0.9938	40	1.6300	0.3908
1.2025	1.9876	80	1.0530	0.5290
0.9391	2.9814	120	0.8234	0.5997
0.858	4.0	161	0.7816	0.6179
0.7816	4.9938	201	0.7162	0.6328
0.7094	5.9876	241	0.6275	0.6936
0.7139	6.9814	281	0.5860	0.7203
0.6836	8.0	322	0.6788	0.6777
0.7075	8.9938	362	0.6239	0.7135
0.6309	9.9876	402	0.6282	0.6995
0.6596	10.9814	442	0.6014	0.6995
0.6589	12.0	483	0.5762	0.7208
0.5947	12.9938	523	0.5150	0.7588
0.5929	13.9876	563	0.5373	0.7403
0.5775	14.9814	603	0.5327	0.7584
0.5465	16.0	644	0.5778	0.7421
0.5614	16.9938	684	0.4826	0.7747
0.5535	17.9876	724	0.4898	0.7788
0.4965	18.9814	764	0.5204	0.7539
0.5177	20.0	805	0.4869	0.7693
0.513	20.9938	845	0.4848	0.7797
0.4824	21.9876	885	0.4821	0.7788
0.5241	22.9814	925	0.4995	0.7715
0.4363	24.0	966	0.5120	0.7806
0.4549	24.9938	1006	0.5031	0.7811
0.4445	25.9876	1046	0.5340	0.7665
0.4337	26.9814	1086	0.5099	0.7783
0.4272	28.0	1127	0.5166	0.7824
0.3732	28.9938	1167	0.5644	0.7620
0.4252	29.9876	1207	0.5171	0.7856
0.3844	30.9814	1247	0.5678	0.7756
0.3663	32.0	1288	0.5667	0.7743
0.3456	32.9938	1328	0.5701	0.7688
0.3809	33.9876	1368	0.5687	0.7729
0.3499	34.9814	1408	0.5615	0.7820
0.3443	36.0	1449	0.5977	0.7806
0.2993	36.9938	1489	0.6292	0.7806
0.3333	37.9876	1529	0.6492	0.7715
0.3586	38.9814	1569	0.6130	0.7756
0.2979	40.0	1610	0.5870	0.7806
0.266	40.9938	1650	0.6225	0.7833
0.2585	41.9876	1690	0.6603	0.7765
0.2741	42.9814	1730	0.6642	0.7752
0.2674	44.0	1771	0.6706	0.7851
0.2619	44.9938	1811	0.6730	0.7715
0.252	45.9876	1851	0.7346	0.7811
0.2417	46.9814	1891	0.6707	0.7829
0.2406	48.0	1932	0.6497	0.7833
0.2348	48.9938	1972	0.6786	0.7833
0.2265	49.9876	2012	0.7158	0.7906
0.1967	50.9814	2052	0.7403	0.7947
0.1932	52.0	2093	0.7282	0.7869
0.2047	52.9938	2133	0.6987	0.7842
0.1966	53.9876	2173	0.7779	0.7851
0.1824	54.9814	2213	0.7815	0.7910
0.1963	56.0	2254	0.6768	0.7933
0.1984	56.9938	2294	0.7527	0.7833
0.1777	57.9876	2334	0.7672	0.7865
0.1666	58.9814	2374	0.7881	0.7901
0.1649	60.0	2415	0.7903	0.7856
0.1785	60.9938	2455	0.7483	0.7874
0.167	61.9876	2495	0.7278	0.7915
0.1514	62.9814	2535	0.8130	0.7761
0.1423	64.0	2576	0.8144	0.7829
0.1476	64.9938	2616	0.8000	0.7815
0.1742	65.9876	2656	0.7660	0.7815
0.1362	66.9814	2696	0.8117	0.7888
0.126	68.0	2737	0.8394	0.7874
0.1278	68.9938	2777	0.8493	0.7847
0.1181	69.9876	2817	0.7959	0.7937
0.1457	70.9814	2857	0.8036	0.7860
0.1322	72.0	2898	0.8474	0.8001
0.1312	72.9938	2938	0.8026	0.7774
0.1146	73.9876	2978	0.8388	0.8064
0.141	74.9814	3018	0.8053	0.7987
0.1396	76.0	3059	0.8439	0.7937
0.113	76.9938	3099	0.9004	0.7919
0.1219	77.9876	3139	0.8423	0.7951
0.1132	78.9814	3179	0.8309	0.7937
0.119	80.0	3220	0.8210	0.8015
0.111	80.9938	3260	0.8238	0.7983
0.0973	81.9876	3300	0.8422	0.7983
0.1118	82.9814	3340	0.8389	0.8010
0.1296	84.0	3381	0.8178	0.8019
0.089	84.9938	3421	0.8456	0.7987
0.1003	85.9876	3461	0.8626	0.8001
0.1123	86.9814	3501	0.8494	0.7928
0.1038	88.0	3542	0.8584	0.8064
0.1055	88.9938	3582	0.8513	0.7933
0.1031	89.9876	3622	0.8592	0.7978
0.1028	90.9814	3662	0.8452	0.7969
0.0998	92.0	3703	0.8605	0.7983
0.1005	92.9938	3743	0.8805	0.7947
0.0936	93.9876	3783	0.8735	0.7969
0.0779	94.9814	3823	0.8776	0.7960
0.0972	96.0	3864	0.8784	0.7951
0.0973	96.9938	3904	0.8782	0.7933
0.0932	97.9876	3944	0.8779	0.7924
0.0863	98.9814	3984	0.8741	0.7919
0.0827	99.3789	4000	0.8737	0.7937

Framework versions

Transformers 4.44.0
Pytorch 1.12.1+cu113
Datasets 2.21.0
Tokenizers 0.19.1