vit_focus / README.md

Fin de l'entraînement

062237a verified 12 months ago

3.22 kB

library_name: transformers
tags:
  - generated_from_trainer
model-index:
  - name: vit_focus
    results: []

vit_focus

This model is a fine-tuned version of 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: 5e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 32
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
num_epochs: 30

Training Loss	Epoch	Step	Validation Loss	Mse	Mae
No log	1.0	10	0.0993	0.1529	0.3380
No log	2.0	20	0.1050	0.1554	0.3409
No log	3.0	30	0.0997	0.1493	0.3352
0.313	4.0	40	0.0656	0.1345	0.3157
0.313	5.0	50	0.0659	0.1366	0.3203
0.313	6.0	60	0.0639	0.1296	0.3119
0.313	7.0	70	0.0639	0.1351	0.3178
0.1742	8.0	80	0.0639	0.1274	0.3086
0.1742	9.0	90	0.0728	0.1294	0.3096
0.1742	10.0	100	0.0671	0.1330	0.3150
0.1742	11.0	110	0.0670	0.1260	0.3067
0.1284	12.0	120	0.0658	0.1253	0.3059
0.1284	13.0	130	0.0641	0.1281	0.3104
0.1284	14.0	140	0.0643	0.1289	0.3105
0.1284	15.0	150	0.0649	0.1342	0.3172
0.0981	16.0	160	0.0656	0.1276	0.3085
0.0981	17.0	170	0.0627	0.1316	0.3136
0.0981	18.0	180	0.0620	0.1343	0.3169
0.0981	19.0	190	0.0632	0.1311	0.3129
0.0767	20.0	200	0.0630	0.1326	0.3143
0.0767	21.0	210	0.0641	0.1299	0.3117
0.0767	22.0	220	0.0634	0.1294	0.3114
0.0767	23.0	230	0.0629	0.1315	0.3130
0.0615	24.0	240	0.0612	0.1308	0.3124
0.0615	25.0	250	0.0599	0.1302	0.3118
0.0615	26.0	260	0.0609	0.1313	0.3132
0.0615	27.0	270	0.0609	0.1301	0.3117