fine_tuned_t5_small_model

This model is a fine-tuned version of t5-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: 2e-05
train_batch_size: 16
eval_batch_size: 16
seed: 42
optimizer: Use 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: 20
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	6	10.6516	0.065	0.0252	0.0587	0.0586	18.5714
No log	2.0	12	7.0831	0.0734	0.0231	0.0642	0.0641	18.4286
No log	3.0	18	5.2606	0.0769	0.0249	0.0655	0.0656	18.3333
8.4329	4.0	24	4.5711	0.0834	0.03	0.0713	0.0714	18.381
8.4329	5.0	30	4.2729	0.089	0.0321	0.0752	0.0751	18.0952
8.4329	6.0	36	4.0881	0.0923	0.0389	0.077	0.0771	18.0952
4.5969	7.0	42	3.9556	0.0861	0.0348	0.0735	0.0732	17.9524
4.5969	8.0	48	3.8522	0.0836	0.0315	0.0709	0.0701	17.8095
4.5969	9.0	54	3.7762	0.0822	0.0304	0.0696	0.0691	17.8095
4.0333	10.0	60	3.7227	0.0799	0.0276	0.0667	0.0665	18.4762
4.0333	11.0	66	3.6849	0.0721	0.0249	0.0609	0.0607	18.619
4.0333	12.0	72	3.6571	0.0745	0.028	0.0639	0.0638	18.6667
4.0333	13.0	78	3.6349	0.075	0.0287	0.0649	0.0648	18.7143
3.8343	14.0	84	3.6187	0.0761	0.0307	0.0661	0.0658	18.7143
3.8343	15.0	90	3.6054	0.0764	0.0309	0.0659	0.0655	18.8095
3.8343	16.0	96	3.5949	0.0795	0.0322	0.0687	0.0684	18.8571
3.7878	17.0	102	3.5866	0.0873	0.0402	0.0764	0.076	18.8095
3.7878	18.0	108	3.5803	0.0873	0.0402	0.0764	0.076	18.8095
3.7878	19.0	114	3.5764	0.0863	0.0392	0.0753	0.075	18.8095
3.7464	20.0	120	3.5741	0.0875	0.0392	0.0762	0.0759	18.8571