working

This model is a fine-tuned version of csebuetnlp/banglat5 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0912

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: 0.001
• train_batch_size: 6
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 48
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine_with_restarts
• lr_scheduler_warmup_steps: 100
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss
39.4442	1.0540	100	0.5396
0.6861	2.1080	200	0.3455
0.4661	3.1621	300	0.3372
0.4512	4.2161	400	0.3265
0.4309	5.2701	500	0.3141
0.4114	6.3241	600	0.2886
0.3679	7.3781	700	0.1776
0.208	8.4321	800	0.1162
0.4885	9.4862	900	0.1208
0.137	10.5402	1000	0.1012
0.1013	11.5942	1100	0.0948
0.0838	12.6482	1200	0.0894
0.0704	13.7022	1300	0.0878
0.0618	14.7563	1400	0.0902
0.0546	15.8103	1500	0.0880
0.049	16.8643	1600	0.0901
0.0443	17.9183	1700	0.0900
0.0417	18.9723	1800	0.0905
0.0396	20.0264	1900	0.0909
0.0373	21.0804	2000	0.0909
0.0369	22.1344	2100	0.0913
0.0358	23.1884	2200	0.0911
0.0356	24.2424	2300	0.0912

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0
• Datasets 2.21.0
• Tokenizers 0.19.1