loha_fine_tuned_cb_croslo

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

• Loss: 1.2436
• Accuracy: 0.3182
• F1: 0.1536

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• training_steps: 400

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.9874	3.5714	50	1.1351	0.3182	0.1591
0.8665	7.1429	100	1.1589	0.3182	0.1536
0.8359	10.7143	150	1.1890	0.3182	0.1536
0.7662	14.2857	200	1.2116	0.3182	0.1536
0.769	17.8571	250	1.2287	0.3182	0.1536
0.7534	21.4286	300	1.2380	0.3182	0.1536
0.7359	25.0	350	1.2421	0.3182	0.1536
0.7449	28.5714	400	1.2436	0.3182	0.1536

Framework versions

• PEFT 0.10.1.dev0
• Transformers 4.40.1
• Pytorch 2.3.0
• Datasets 2.19.0
• Tokenizers 0.19.1