robeczech_lr1e-05_bs16_train287

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

• Loss: 0.2219
• Precision: 0.9364
• Recall: 0.9516
• F1: 0.9439
• Accuracy: 0.9681

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: 1e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	18	1.6464	0.0	0.0	0.0	0.5666
No log	2.0	36	1.3151	1.0	0.0005	0.0010	0.5668
No log	3.0	54	1.1082	0.3172	0.2849	0.3002	0.6802
No log	4.0	72	0.9408	0.5488	0.5292	0.5388	0.7882
No log	5.0	90	0.7478	0.6970	0.6953	0.6962	0.8571
No log	6.0	108	0.5949	0.7631	0.7716	0.7673	0.8895
No log	7.0	126	0.4931	0.8374	0.8479	0.8426	0.9221
No log	8.0	144	0.4231	0.8736	0.8846	0.8791	0.9393
No log	9.0	162	0.3772	0.8838	0.8923	0.8880	0.9437
No log	10.0	180	0.3473	0.8852	0.8938	0.8895	0.9447
No log	11.0	198	0.3302	0.8907	0.8972	0.8939	0.9473
No log	12.0	216	0.3165	0.8893	0.9000	0.8946	0.9466
No log	13.0	234	0.3041	0.8912	0.8976	0.8944	0.9473
No log	14.0	252	0.2940	0.8946	0.8972	0.8959	0.9483
No log	15.0	270	0.2898	0.8936	0.9005	0.8971	0.9481
No log	16.0	288	0.2823	0.8924	0.9010	0.8967	0.9481
No log	17.0	306	0.2777	0.8941	0.9010	0.8975	0.9487
No log	18.0	324	0.2717	0.8972	0.9020	0.8996	0.9500
No log	19.0	342	0.2644	0.9034	0.9078	0.9056	0.9523
No log	20.0	360	0.2585	0.9051	0.9116	0.9083	0.9535
No log	21.0	378	0.2523	0.9131	0.9184	0.9157	0.9567
No log	22.0	396	0.2523	0.9136	0.9189	0.9162	0.9569
No log	23.0	414	0.2473	0.9177	0.9256	0.9216	0.9590
No log	24.0	432	0.2476	0.9186	0.9261	0.9223	0.9592
No log	25.0	450	0.2443	0.9198	0.9252	0.9225	0.9594
No log	26.0	468	0.2439	0.9189	0.9247	0.9218	0.9592
No log	27.0	486	0.2407	0.9203	0.9256	0.9230	0.9596
0.4988	28.0	504	0.2393	0.9207	0.9252	0.9229	0.9598
0.4988	29.0	522	0.2397	0.9204	0.9266	0.9235	0.9598
0.4988	30.0	540	0.2392	0.9214	0.9281	0.9247	0.9604

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1