stage2

This model is a fine-tuned version of linker_v4/stage1 on the jarod0411/linker_v2 dataset. It achieves the following results on the evaluation set:

• Loss: 0.3246
• Accuracy: 0.8954

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: 5e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 1
• distributed_type: multi-GPU
• num_devices: 5
• total_train_batch_size: 80
• total_eval_batch_size: 80
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.3608	1.0	43075	0.3509	0.8883
0.3495	2.0	86150	0.3408	0.8910
0.3435	3.0	129225	0.3355	0.8925
0.3394	4.0	172300	0.3322	0.8933
0.3367	5.0	215375	0.3298	0.8940
0.3343	6.0	258450	0.3279	0.8945
0.3326	7.0	301525	0.3267	0.8948
0.3311	8.0	344600	0.3256	0.8951
0.33	9.0	387675	0.3249	0.8953
0.3296	10.0	430750	0.3246	0.8954

Framework versions

• Transformers 4.38.0.dev0
• Pytorch 2.2.0+cu121
• Datasets 2.17.0
• Tokenizers 0.15.2