xlm-roberta-base_amh_corr_0.0001

This model is a fine-tuned version of FacebookAI/xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0466
• Spearman Corr: 0.0096

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: 32
• eval_batch_size: 128
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Spearman Corr
No log	0.9	200	0.0461	nan
No log	1.81	400	0.0467	nan
0.051	2.71	600	0.0464	-0.0412
0.051	3.62	800	0.0462	0.0202
0.0506	4.52	1000	0.0469	nan
0.0506	5.43	1200	0.0462	-0.0169
0.0498	6.33	1400	0.0469	nan
0.0498	7.24	1600	0.0462	nan
0.0499	8.14	1800	0.0466	0.0096

Framework versions

• Transformers 4.37.2
• Pytorch 2.2.0+cu121
• Datasets 2.17.0
• Tokenizers 0.15.2