This model is a fine-tuned version of 0x-Jayveersinh-Raj/fabric_classifier on the arrow dataset.
It achieves the following results on the evaluation set:
Loss: 2.0567
Accuracy: 0.65
F1 Macro: 0.2581
F1 Micro: 0.5
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: 3e-05
train_batch_size: 8
eval_batch_size: 16
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
