nci-scorer-nci-only

This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0198

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: 32
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.3145	100	0.0936
No log	0.6289	200	0.0437
No log	0.9434	300	0.0314
No log	1.2579	400	0.0271
0.0683	1.5723	500	0.0268
0.0683	1.8868	600	0.0257
0.0683	2.2013	700	0.0232
0.0683	2.5157	800	0.0228
0.0683	2.8302	900	0.0218
0.0252	3.1447	1000	0.0226
0.0252	3.4591	1100	0.0221
0.0252	3.7736	1200	0.0217
0.0252	4.0881	1300	0.0220
0.0252	4.4025	1400	0.0207
0.0197	4.7170	1500	0.0215
0.0197	5.0314	1600	0.0207
0.0197	5.3459	1700	0.0202
0.0197	5.6604	1800	0.0208
0.0197	5.9748	1900	0.0200
0.0175	6.2893	2000	0.0199
0.0175	6.6038	2100	0.0200
0.0175	6.9182	2200	0.0197
0.0175	7.2327	2300	0.0199
0.0175	7.5472	2400	0.0197
0.0156	7.8616	2500	0.0197
0.0156	8.1761	2600	0.0198
0.0156	8.4906	2700	0.0196
0.0156	8.8050	2800	0.0196
0.0156	9.1195	2900	0.0195
0.0144	9.4340	3000	0.0194
0.0144	9.7484	3100	0.0195

Framework versions

• Transformers 4.57.3
• Pytorch 2.9.1+cu128
• Datasets 4.4.1
• Tokenizers 0.22.1