lilt_robeczech_lr2e-05_bs4_train150

This model is a fine-tuned version of xkaska02/lilt-robeczech-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.5876
• Precision: 0.6770
• Recall: 0.6553
• F1: 0.6660
• Accuracy: 0.8160

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: 8
• eval_batch_size: 8
• 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	75	1.2417	0.6929	0.1479	0.2438	0.5845
No log	2.0	150	1.0821	0.5272	0.3747	0.4380	0.6083
No log	3.0	225	1.0111	0.5588	0.3864	0.4569	0.6223
No log	4.0	300	1.0223	0.6159	0.3040	0.4071	0.6206
No log	5.0	375	1.0028	0.5296	0.4316	0.4756	0.6187
No log	6.0	450	0.9913	0.5352	0.4138	0.4667	0.6182
1.0363	7.0	525	0.9728	0.5436	0.4438	0.4887	0.6315
1.0363	8.0	600	0.7223	0.6495	0.5775	0.6114	0.7593
1.0363	9.0	675	0.6335	0.6583	0.6299	0.6438	0.7940
1.0363	10.0	750	0.6162	0.6759	0.6299	0.6521	0.8027
1.0363	11.0	825	0.6024	0.6670	0.6406	0.6535	0.8053
1.0363	12.0	900	0.6027	0.6647	0.6421	0.6532	0.8079
1.0363	13.0	975	0.5876	0.6770	0.6553	0.6660	0.8160
0.4057	14.0	1050	0.5959	0.6698	0.6528	0.6612	0.8136
0.4057	15.0	1125	0.5906	0.6831	0.6543	0.6684	0.8179
0.4057	16.0	1200	0.5994	0.6801	0.6497	0.6646	0.8169

Framework versions

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