my_awesome_ne5_model

This model is a fine-tuned version of cointegrated/rubert-tiny2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.2628
• Precision: 0.4231
• Recall: 0.4158
• F1: 0.4194
• Accuracy: 0.9189

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: 0.0001
• train_batch_size: 64
• eval_batch_size: 64
• 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: reduce_lr_on_plateau
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	13	0.6601	0.0	0.0	0.0	0.8659
No log	2.0	26	0.5417	0.0	0.0	0.0	0.8659
No log	3.0	39	0.4561	0.2071	0.1344	0.1630	0.8793
No log	4.0	52	0.3994	0.2222	0.1694	0.1922	0.8865
No log	5.0	65	0.3595	0.2644	0.2249	0.2430	0.8943
No log	6.0	78	0.3278	0.3112	0.2639	0.2856	0.9018
No log	7.0	91	0.3087	0.3406	0.2976	0.3177	0.9058
No log	8.0	104	0.2945	0.3438	0.3238	0.3335	0.9075
No log	9.0	117	0.2881	0.3519	0.3470	0.3494	0.9079
No log	10.0	130	0.2786	0.3760	0.3696	0.3728	0.9115
No log	11.0	143	0.2732	0.3912	0.3809	0.3860	0.9136
No log	12.0	156	0.2674	0.4104	0.4030	0.4066	0.9159
No log	13.0	169	0.2686	0.4013	0.4045	0.4029	0.9153
No log	14.0	182	0.2628	0.4231	0.4158	0.4194	0.9189
No log	15.0	195	0.2666	0.4278	0.4207	0.4242	0.9187
No log	16.0	208	0.2702	0.4249	0.4264	0.4257	0.9182
No log	17.0	221	0.2707	0.4319	0.4428	0.4373	0.9186
No log	18.0	234	0.2761	0.4108	0.4284	0.4194	0.9158

Framework versions

• Transformers 4.51.1
• Pytorch 2.5.1+cu124
• Datasets 3.5.0
• Tokenizers 0.21.0