ania3000/ossbert-morph-e-v2

trainer_output

This model is a fine-tuned version of AlexeySorokin/ossbert-onc-unlab-bs64-5epochs on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3650
• Accuracy: 95.6576
• Sentence accuracy: 59.4495

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: 8
• eval_batch_size: 8
• 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
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Sentence accuracy
1.0698	1.0	546	0.3788	92.1082	43.1193
0.3296	2.0	1092	0.2647	94.0969	51.9266
0.2081	3.0	1638	0.2419	94.4745	52.6606
0.1458	4.0	2184	0.2334	94.9906	53.9450
0.1071	5.0	2730	0.2254	95.1290	55.7798
0.0806	6.0	3276	0.2527	95.1919	57.7982
0.0599	7.0	3822	0.2556	95.3052	56.6972
0.042	8.0	4368	0.2634	95.3556	57.4312
0.0343	9.0	4914	0.2695	95.4940	57.6147
0.0237	10.0	5460	0.2970	95.3304	57.9817
0.0152	11.0	6006	0.2884	95.5444	58.7156
0.0102	12.0	6552	0.2950	95.5318	58.1651
0.0088	13.0	7098	0.3069	95.6199	57.4312
0.0049	14.0	7644	0.3217	95.7080	58.7156
0.0056	15.0	8190	0.3223	95.6576	58.8991
0.0044	16.0	8736	0.3220	95.7709	58.5321
0.0036	17.0	9282	0.3391	95.7206	59.2661
0.0035	18.0	9828	0.3515	95.5066	58.3486
0.0035	19.0	10374	0.3600	95.4940	58.3486
0.0022	20.0	10920	0.3650	95.6576	59.4495

Framework versions

• Transformers 4.57.3
• Pytorch 2.10.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.2