bt5_peft_test

This model is a fine-tuned version of csebuetnlp/banglat5 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 8.7936

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: 0.0001
• 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
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
14.4734	0.05	10	10.4935
13.7539	0.1	20	10.7648
15.0363	0.15	30	10.5645
12.407	0.2	40	10.4939
13.9381	0.25	50	10.3898
13.7289	0.3	60	10.4080
14.3071	0.35	70	9.9145
12.541	0.4	80	9.9789
12.1909	0.45	90	9.4685
12.0731	0.5	100	9.1319
12.15	0.54	110	8.9585
11.2228	0.59	120	8.9887
13.1212	0.64	130	8.8966
11.3415	0.69	140	8.8654
12.0359	0.74	150	8.9607
11.5232	0.79	160	8.9678
11.2903	0.84	170	9.0175
10.1588	0.89	180	8.9056
10.4939	0.94	190	8.8735
9.914	0.99	200	8.7936

Framework versions

• PEFT 0.11.1
• Transformers 4.39.3
• Pytorch 2.1.2
• Datasets 2.18.0
• Tokenizers 0.15.2