README.md

metadata

base_model: MBZUAI/swiftformer-xs
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: swiftformer-xs-ve-U13-b-80
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: default
          split: validation
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.8695652173913043

swiftformer-xs-ve-U13-b-80

This model is a fine-tuned version of MBZUAI/swiftformer-xs on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.6163
• Accuracy: 0.8696

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: 0.0002
• 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: 80

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.92	6	1.4461	0.1087
1.3993	2.0	13	1.4435	0.1087
1.3993	2.92	19	1.4389	0.1087
1.3849	4.0	26	1.4284	0.1087
1.3287	4.92	32	1.4223	0.1304
1.3287	6.0	39	1.4647	0.1087
1.2128	6.92	45	1.4184	0.1739
1.122	8.0	52	1.3262	0.1957
1.122	8.92	58	1.3298	0.1957
1.0062	10.0	65	1.2035	0.3696
0.872	10.92	71	1.3667	0.3261
0.872	12.0	78	1.2013	0.4348
0.814	12.92	84	0.9996	0.6957
0.7228	14.0	91	0.9706	0.7609
0.7228	14.92	97	0.9295	0.7391
0.6473	16.0	104	0.8988	0.7174
0.5696	16.92	110	0.9612	0.6739
0.5696	18.0	117	0.8713	0.7609
0.5546	18.92	123	0.8050	0.7609
0.4747	20.0	130	0.7725	0.7609
0.4747	20.92	136	0.7933	0.6957
0.4393	22.0	143	0.7665	0.6957
0.4393	22.92	149	0.7886	0.7174
0.4077	24.0	156	0.7824	0.7391
0.3326	24.92	162	0.7021	0.7391
0.3326	26.0	169	0.6346	0.8261
0.315	26.92	175	0.6163	0.8696
0.2729	28.0	182	0.6938	0.8043
0.2729	28.92	188	0.7417	0.8043
0.2218	30.0	195	0.6669	0.7826
0.2499	30.92	201	0.7111	0.7174
0.2499	32.0	208	0.6730	0.7826
0.2218	32.92	214	0.6512	0.8043
0.2037	34.0	221	0.7165	0.7391
0.2037	34.92	227	0.6300	0.8261
0.2367	36.0	234	0.7421	0.7826
0.1835	36.92	240	0.6644	0.8043
0.1835	38.0	247	0.6251	0.8261
0.2073	38.92	253	0.6431	0.7826
0.1643	40.0	260	0.6348	0.8043
0.1643	40.92	266	0.6192	0.8043
0.1685	42.0	273	0.6753	0.8043
0.1685	42.92	279	0.7440	0.7826
0.1539	44.0	286	0.7505	0.8043
0.1658	44.92	292	0.6331	0.8261
0.1658	46.0	299	0.6550	0.8043
0.1596	46.92	305	0.6824	0.8043
0.1534	48.0	312	0.6971	0.8043
0.1534	48.92	318	0.6347	0.8261
0.1677	50.0	325	0.6392	0.8261
0.1453	50.92	331	0.6369	0.8043
0.1453	52.0	338	0.6230	0.7826
0.1385	52.92	344	0.6432	0.7609
0.1221	54.0	351	0.6757	0.7391
0.1221	54.92	357	0.7383	0.7609
0.1433	56.0	364	0.7100	0.7609
0.1567	56.92	370	0.6862	0.7609
0.1567	58.0	377	0.6654	0.7609
0.1361	58.92	383	0.6665	0.7826
0.1157	60.0	390	0.6439	0.7826
0.1157	60.92	396	0.6306	0.8261
0.0934	62.0	403	0.6546	0.8043
0.0934	62.92	409	0.6651	0.8043
0.1123	64.0	416	0.6568	0.7609
0.0855	64.92	422	0.6507	0.7609
0.0855	66.0	429	0.6667	0.8043
0.1135	66.92	435	0.6516	0.7826
0.0932	68.0	442	0.6596	0.8043
0.0932	68.92	448	0.6772	0.8043
0.1228	70.0	455	0.6526	0.7609
0.0878	70.92	461	0.6731	0.8261
0.0878	72.0	468	0.6351	0.7826
0.1073	72.92	474	0.6269	0.7826
0.1028	73.85	480	0.6743	0.7826

Framework versions

• Transformers 4.36.2
• Pytorch 2.1.2+cu118
• Datasets 2.16.1
• Tokenizers 0.15.0