rubert-tiny2-kinopoisk

This model is a fine-tuned version of cointegrated/rubert-tiny2 on Glepka/kinopoisk_classification. It achieves the following results on the evaluation set:

• Loss: 0.7214
• Accuracy: 0.738

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: 64
• eval_batch_size: 64
• 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: 12

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.7896	1.0	891	0.5692	0.7367
0.5548	2.0	1782	0.5882	0.734
0.5207	3.0	2673	0.5731	0.7467
0.4806	4.0	3564	0.5806	0.7487
0.4615	5.0	4455	0.6030	0.746
0.4206	6.0	5346	0.6197	0.7453
0.3913	7.0	6237	0.6427	0.7413
0.3678	8.0	7128	0.6605	0.7413
0.3392	9.0	8019	0.6922	0.7367
0.3299	10.0	8910	0.7000	0.738
0.3125	11.0	9801	0.7139	0.736
0.308	12.0	10692	0.7214	0.738

Framework versions

• Transformers 4.57.1
• Pytorch 2.8.0+cu126
• Datasets 4.4.1
• Tokenizers 0.22.1