robeczech_lr2e-05_bs4_train5

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: 1.1314
• Precision: 0.5641
• Recall: 0.3322
• F1: 0.4181
• Accuracy: 0.7120

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: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Use 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	2	2.3031	0.0870	0.0068	0.0125	0.5628
No log	2.0	4	2.0883	0.0	0.0	0.0	0.5666
No log	3.0	6	1.9029	0.0	0.0	0.0	0.5666
No log	4.0	8	1.7831	0.0	0.0	0.0	0.5666
No log	5.0	10	1.7223	0.0	0.0	0.0	0.5666
No log	6.0	12	1.6812	0.0	0.0	0.0	0.5666
No log	7.0	14	1.6437	0.0	0.0	0.0	0.5666
No log	8.0	16	1.6091	0.0	0.0	0.0	0.5666
No log	9.0	18	1.5787	0.0	0.0	0.0	0.5666
No log	10.0	20	1.5511	0.0	0.0	0.0	0.5666
No log	11.0	22	1.5191	0.0	0.0	0.0	0.5666
No log	12.0	24	1.4842	0.0	0.0	0.0	0.5666
No log	13.0	26	1.4505	0.5	0.0005	0.0010	0.5668
No log	14.0	28	1.4157	0.6522	0.0072	0.0143	0.5697
No log	15.0	30	1.3825	0.54	0.0130	0.0255	0.5722
No log	16.0	32	1.3515	0.6047	0.0251	0.0482	0.5774
No log	17.0	34	1.3216	0.6333	0.0459	0.0855	0.5862
No log	18.0	36	1.2934	0.6038	0.0773	0.1370	0.5996
No log	19.0	38	1.2691	0.5943	0.1202	0.2	0.6183
No log	20.0	40	1.2459	0.5925	0.1685	0.2624	0.6388
No log	21.0	42	1.2255	0.6032	0.2202	0.3226	0.6601
No log	22.0	44	1.2094	0.5925	0.2380	0.3396	0.6679
No log	23.0	46	1.1961	0.5969	0.2603	0.3625	0.6773
No log	24.0	48	1.1865	0.5931	0.2646	0.3659	0.6792
No log	25.0	50	1.1778	0.5944	0.2767	0.3776	0.6840
No log	26.0	52	1.1686	0.6022	0.2931	0.3943	0.6909
No log	27.0	54	1.1594	0.5837	0.3047	0.4004	0.6957
No log	28.0	56	1.1532	0.5717	0.3100	0.4020	0.6978
No log	29.0	58	1.1496	0.5688	0.3153	0.4057	0.6995
No log	30.0	60	1.1481	0.5654	0.3192	0.4080	0.7007

Framework versions

• Transformers 4.51.3
• Pytorch 2.7.0+cu126
• Datasets 3.6.0
• Tokenizers 0.21.1