SOUMYADEEPSAR/mbert_CoLI_dravidian_kannada

results_new

This model is a fine-tuned version of bert-base-multilingual-cased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.2322

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: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	234	0.2011
No log	2.0	468	0.2151
0.246	3.0	702	0.1992
0.246	4.0	936	0.1964
0.0694	5.0	1170	0.1924
0.0694	6.0	1404	0.2061
0.028	7.0	1638	0.2185
0.028	8.0	1872	0.2135
0.0155	9.0	2106	0.2338
0.0155	10.0	2340	0.2322

Framework versions

• Transformers 4.42.4
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1