ania3000/ossbert-morph-2-1

trainer_output

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

• Loss: 0.3130
• Accuracy: 95.8657
• Sentence accuracy: 60.1835

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
1.0267	1.0	546	0.3459	92.2075	44.2202
0.3155	2.0	1092	0.2568	94.1118	51.5596
0.201	3.0	1638	0.2254	94.8259	56.6972
0.142	4.0	2184	0.2280	94.9261	55.9633
0.1103	5.0	2730	0.2106	95.5149	58.1651
0.084	6.0	3276	0.2431	95.2393	55.9633
0.0652	7.0	3822	0.2433	95.5149	59.0826
0.0495	8.0	4368	0.2505	95.5149	57.7982
0.039	9.0	4914	0.2568	95.7404	59.6330
0.0288	10.0	5460	0.2769	95.7279	60.1835
0.0191	11.0	6006	0.2743	95.6527	59.4495
0.0132	12.0	6552	0.2876	95.7154	58.8991
0.0119	13.0	7098	0.2814	95.8281	59.4495
0.0094	14.0	7644	0.2910	95.8156	60.9174
0.0071	15.0	8190	0.2983	96.0035	60.3670
0.0049	16.0	8736	0.3077	95.9409	59.4495
0.0053	17.0	9282	0.3148	95.9158	60.7339
0.0048	18.0	9828	0.3096	95.8908	61.1009
0.0033	19.0	10374	0.3130	95.8657	60.1835

Framework versions

• Transformers 4.57.3
• Pytorch 2.11.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.2