xlm-roberta-base_lr0.0001_seed42_basic_original_amh-hau-eng_train

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: 0.0468
• Spearman Corr: 0.0355

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	1.55	200	0.0533	-0.1261
0.0648	3.1	400	0.0450	0.0883
0.0593	4.65	600	0.0551	nan
0.0584	6.2	800	0.0453	0.0647
0.0584	7.75	1000	0.0535	0.1302
0.0584	9.3	1200	0.0447	0.1247
0.0564	10.85	1400	0.0568	0.0145
0.0564	12.4	1600	0.0447	nan
0.056	13.95	1800	0.0491	0.0365
0.056	15.5	2000	0.0472	nan
0.0564	17.05	2200	0.0473	nan
0.0555	18.6	2400	0.0457	nan
0.0555	20.16	2600	0.0472	-0.0083
0.0555	21.71	2800	0.0473	nan
0.055	23.26	3000	0.0447	0.0283
0.0549	24.81	3200	0.0468	0.0355

Framework versions

• Transformers 4.37.2
• Pytorch 2.1.0+cu121
• Datasets 2.17.0
• Tokenizers 0.15.2