README.md

metadata

license: apache-2.0
base_model: microsoft/cvt-21-384-22k
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: cvt-21-384-22k-finetuned-LeafType
    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: 0.9310344827586207

cvt-21-384-22k-finetuned-LeafType

This model is a fine-tuned version of microsoft/cvt-21-384-22k on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.2232
• Accuracy: 0.9310

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: 2e-05
• train_batch_size: 160
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 60

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	11	0.4857	0.7586
No log	2.0	22	0.2275	0.9163
No log	3.0	33	0.2791	0.8867
No log	4.0	44	0.3083	0.8522
No log	5.0	55	0.1496	0.9458
No log	6.0	66	0.2092	0.8966
No log	7.0	77	0.3349	0.8522
No log	8.0	88	0.1513	0.9557
No log	9.0	99	0.1385	0.9557
No log	10.0	110	0.1676	0.9360
No log	11.0	121	0.1284	0.9507
No log	12.0	132	0.1981	0.9015
No log	13.0	143	0.1346	0.9360
No log	14.0	154	0.3235	0.8621
No log	15.0	165	0.1506	0.9310
No log	16.0	176	0.1536	0.9360
No log	17.0	187	0.1529	0.9458
No log	18.0	198	0.2431	0.8966
No log	19.0	209	0.2251	0.9163
No log	20.0	220	0.2068	0.9163
No log	21.0	231	0.1429	0.9458
No log	22.0	242	0.2078	0.9163
No log	23.0	253	0.1428	0.9360
No log	24.0	264	0.2391	0.8916
No log	25.0	275	0.1512	0.9458
No log	26.0	286	0.1512	0.9458
No log	27.0	297	0.1303	0.9606
No log	28.0	308	0.2381	0.8867
No log	29.0	319	0.1544	0.9458
No log	30.0	330	0.2149	0.8916
No log	31.0	341	0.1387	0.9557
No log	32.0	352	0.1698	0.9360
No log	33.0	363	0.1750	0.9409
No log	34.0	374	0.2041	0.9113
No log	35.0	385	0.1888	0.9360
No log	36.0	396	0.1527	0.9458
No log	37.0	407	0.1721	0.9360
No log	38.0	418	0.1720	0.9409
No log	39.0	429	0.1810	0.9360
No log	40.0	440	0.2041	0.9261
No log	41.0	451	0.1945	0.9261
No log	42.0	462	0.2211	0.9163
No log	43.0	473	0.2284	0.9163
No log	44.0	484	0.1905	0.9261
No log	45.0	495	0.2329	0.9113
0.2716	46.0	506	0.1839	0.9360
0.2716	47.0	517	0.1854	0.9310
0.2716	48.0	528	0.1923	0.9310
0.2716	49.0	539	0.2254	0.9261
0.2716	50.0	550	0.2677	0.9015
0.2716	51.0	561	0.2620	0.9015
0.2716	52.0	572	0.2224	0.9261
0.2716	53.0	583	0.2238	0.9310
0.2716	54.0	594	0.2355	0.9212
0.2716	55.0	605	0.2533	0.9261
0.2716	56.0	616	0.2237	0.9310
0.2716	57.0	627	0.2296	0.9310
0.2716	58.0	638	0.2361	0.9310
0.2716	59.0	649	0.2358	0.9310
0.2716	60.0	660	0.2232	0.9310

Framework versions

• Transformers 4.38.1
• Pytorch 1.10.0+cu111
• Datasets 2.17.1
• Tokenizers 0.15.2