metadata

base_model: csebuetnlp/mT5_multilingual_XLSum
tags:
  - generated_from_trainer
model-index:
  - name: HappyNews_1_loadbest
    results: []

HappyNews_1_loadbest

This model is a fine-tuned version of csebuetnlp/mT5_multilingual_XLSum on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 3.1967

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: 5e-05
train_batch_size: 2
eval_batch_size: 2
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 1000
num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss
4.1092	0.29	100	4.0260
4.3545	0.58	200	3.6022
3.818	0.87	300	3.3815
3.2577	1.16	400	3.2590
3.1005	1.45	500	3.1290
3.0309	1.73	600	3.0690
3.0128	2.02	700	3.0172
2.4054	2.31	800	3.0086
2.7848	2.6	900	3.0103
2.4307	2.89	1000	2.9606
2.3408	3.18	1100	2.9490
2.4232	3.47	1200	2.9333
2.5301	3.76	1300	2.9138
1.9984	4.05	1400	2.9422
2.1215	4.34	1500	2.9620
1.859	4.62	1600	2.9550
1.8986	4.91	1700	2.9654
1.847	5.2	1800	3.0660
1.7843	5.49	1900	3.0169
1.9724	5.78	2000	3.0131
1.6603	6.07	2100	3.0816
1.4024	6.36	2200	3.0947
1.2758	6.65	2300	3.0688
1.7435	6.94	2400	3.0203
1.2973	7.23	2500	3.1221
1.282	7.51	2600	3.1566
1.4837	7.8	2700	3.1047
1.6313	8.09	2800	3.1343
1.4611	8.38	2900	3.1634
1.0115	8.67	3000	3.1751
1.4337	8.96	3100	3.1701
1.1845	9.25	3200	3.1881
1.2019	9.54	3300	3.1998
1.1448	9.83	3400	3.1967

Framework versions

Transformers 4.35.2
Pytorch 2.1.0+cu121
Datasets 2.16.1
Tokenizers 0.15.1