TannerGladson/chess-roberta-whole-move-tuned

2024.07.20-19.49

This model is a fine-tuned version of TannerGladson/chess-roberta on the TannerGladson/chess-roberta-whole-move-tuning dataset. It achieves the following results on the evaluation set:

• Loss: 0.2611
• Accuracy: 0.9024

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: 0.0005
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• gradient_accumulation_steps: 16
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.98) and epsilon=1e-06
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 200
• training_steps: 5000

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.5262	0.2485	1000	0.4272	0.8519
0.413	0.4970	2000	0.3650	0.8711
0.3505	0.7455	3000	0.3138	0.8852
0.3111	0.9939	4000	0.2829	0.8950
0.2817	1.2424	5000	0.2596	0.9025

Framework versions

• Transformers 4.42.4
• Pytorch 2.0.1+cu117
• Datasets 2.17.1
• Tokenizers 0.19.1