deberta-v3-large-ft-icar-a-v0.11

This model is a fine-tuned version of microsoft/deberta-v3-large 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: 2e-06
train_batch_size: 1
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 3
total_train_batch_size: 3
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: 50

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
2.6872	1.0	871	0.6573	0.8453	0.8798	0.7874	0.7908
1.9067	2.0	1742	0.7585	0.8806	0.9047	0.8402	0.8553
1.4171	3.0	2613	0.6031	0.9081	0.8979	0.9058	0.9008
1.1236	4.0	3484	0.7067	0.9127	0.9108	0.9003	0.9044
1.0755	5.0	4355	0.8168	0.8974	0.9084	0.8688	0.8839
0.8824	6.0	5226	0.7756	0.9112	0.9031	0.9139	0.9063
0.651	7.0	6097	0.7307	0.9158	0.9165	0.9019	0.9080
0.6468	8.0	6968	0.8609	0.9112	0.9006	0.8949	0.8966
0.4654	9.0	7839	0.8359	0.9096	0.8910	0.9053	0.8974
0.4483	10.0	8710	0.7813	0.9158	0.9107	0.8999	0.9044
0.3246	11.0	9581	0.8845	0.9158	0.9025	0.9096	0.9051
0.2939	12.0	10452	0.8122	0.9204	0.9146	0.9067	0.9100
0.2661	13.0	11323	1.0101	0.9096	0.8986	0.8994	0.8983
0.2874	14.0	12194	0.9044	0.9127	0.9043	0.9057	0.9038
0.2304	15.0	13065	1.0021	0.9066	0.8940	0.9120	0.8988
0.1991	16.0	13936	0.8841	0.9158	0.9052	0.9034	0.9036
0.1957	17.0	14807	1.0139	0.9096	0.9035	0.9025	0.9013
0.116	18.0	15678	0.9821	0.9127	0.9049	0.9042	0.9027