mt5-small-asqp-rest16

This model is a fine-tuned version of google/mt5-small on the None 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: 0.0003
train_batch_size: 16
eval_batch_size: 16
seed: 42
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: 15

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1
No log	1.0	79	1.7680	0.0	0.0	0
No log	2.0	158	0.7157	0.0877	0.0986	0.0929
No log	3.0	237	0.5655	0.0915	0.2071	0.127
No log	4.0	316	0.4808	0.2169	0.2682	0.2399
No log	5.0	395	0.4643	0.16	0.3412	0.2179
No log	6.0	474	0.4340	0.2551	0.3669	0.301
2.1045	7.0	553	0.4385	0.2283	0.3787	0.2849
2.1045	8.0	632	0.4229	0.2456	0.3846	0.2998
2.1045	9.0	711	0.4244	0.317	0.3945	0.3515
2.1045	10.0	790	0.4190	0.3296	0.4024	0.3623
2.1045	11.0	869	0.4211	0.3243	0.4004	0.3583
2.1045	12.0	948	0.4274	0.3069	0.4122	0.3519
0.3564	13.0	1027	0.4162	0.3209	0.4083	0.3594
0.3564	14.0	1106	0.4131	0.3365	0.4221	0.3745
0.3564	15.0	1185	0.4146	0.3287	0.4221	0.3696