uitviquad_noseg_bart

This model is a fine-tuned version of google/mt5-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.7253

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: 1e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss
20.7767	0.45	100	10.0605
13.9887	0.9	200	7.6241
9.9698	1.35	300	5.6805
7.0255	1.8	400	3.2683
5.424	2.25	500	2.6617
4.731	2.7	600	1.9936
4.0452	3.15	700	1.6257
3.4385	3.6	800	1.4585
2.9751	4.05	900	1.3627
2.6369	4.5	1000	1.2824
2.3538	4.95	1100	1.2082
2.1737	5.4	1200	1.1418
2.0271	5.85	1300	1.0817
1.9121	6.3	1400	1.0290
1.8308	6.75	1500	0.9858
1.7694	7.2	1600	0.9456
1.7025	7.65	1700	0.9107
1.6458	8.1	1800	0.8782
1.6022	8.55	1900	0.8516
1.5802	9.0	2000	0.8288
1.5482	9.45	2100	0.8119
1.4982	9.9	2200	0.7938
1.4836	10.35	2300	0.7802
1.4647	10.8	2400	0.7680
1.4437	11.25	2500	0.7571
1.4165	11.7	2600	0.7498
1.4275	12.15	2700	0.7422
1.4045	12.59	2800	0.7375
1.4104	13.04	2900	0.7324
1.366	13.49	3000	0.7296
1.3912	13.94	3100	0.7276
1.3615	14.39	3200	0.7260
1.3801	14.84	3300	0.7253

Framework versions

• Transformers 4.28.0
• Pytorch 2.0.0+cu118
• Datasets 2.11.0
• Tokenizers 0.13.3