xlmr-si-en-no_shuffled-orig-test1000

This model is a fine-tuned version of xlm-roberta-base on the wmt20_mlqe_task1 dataset. It achieves the following results on the evaluation set:

• Loss: 0.6696
• R Squared: -0.0587
• Mae: 0.6388
• Pearson R: 0.5256

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: 5

Training results

Training Loss	Epoch	Step	Validation Loss	R Squared	Mae	Pearson R
No log	1.0	438	0.5300	0.1620	0.6148	0.4337
0.8691	2.0	876	0.4828	0.2366	0.5737	0.5227
0.7034	3.0	1314	0.7226	-0.1425	0.6706	0.5349
0.539	4.0	1752	0.6373	-0.0077	0.6250	0.5361
0.3837	5.0	2190	0.6696	-0.0587	0.6388	0.5256

Framework versions

• Transformers 4.34.1
• Pytorch 2.0.1+cu117
• Datasets 2.14.6
• Tokenizers 0.14.1