--- license: mit base_model: cointegrated/rut5-small tags: - generated_from_trainer model-index: - name: text-normalization-ru-new results: [] --- # text-normalization-ru-new This model is a fine-tuned version of [cointegrated/rut5-small](https://huggingface.co/cointegrated/rut5-small) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.0318 - Mean Distance: 0 - Max Distance: 11 ## 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: 0.001 - train_batch_size: 30 - eval_batch_size: 30 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_ratio: 0.1 - num_epochs: 60 ### Training results | Training Loss | Epoch | Step | Validation Loss | Mean Distance | Max Distance | |:-------------:|:-----:|:------:|:---------------:|:-------------:|:------------:| | 0.2251 | 1.0 | 3334 | 0.1190 | 3 | 29 | | 0.1179 | 2.0 | 6668 | 0.0574 | 2 | 31 | | 0.0848 | 3.0 | 10002 | 0.0436 | 1 | 15 | | 0.0618 | 4.0 | 13336 | 0.0359 | 1 | 20 | | 0.0532 | 5.0 | 16670 | 0.0315 | 0 | 11 | | 0.0446 | 6.0 | 20004 | 0.0299 | 0 | 16 | | 0.0388 | 7.0 | 23338 | 0.0295 | 0 | 15 | | 0.0311 | 8.0 | 26672 | 0.0287 | 0 | 15 | | 0.0269 | 9.0 | 30006 | 0.0241 | 0 | 15 | | 0.0232 | 10.0 | 33340 | 0.0228 | 0 | 13 | | 0.0203 | 11.0 | 36674 | 0.0243 | 0 | 16 | | 0.0173 | 12.0 | 40008 | 0.0250 | 0 | 15 | | 0.0151 | 13.0 | 43342 | 0.0244 | 0 | 9 | | 0.0136 | 14.0 | 46676 | 0.0234 | 0 | 15 | | 0.0123 | 15.0 | 50010 | 0.0221 | 0 | 9 | | 0.0113 | 16.0 | 53344 | 0.0244 | 0 | 12 | | 0.01 | 17.0 | 56678 | 0.0226 | 0 | 13 | | 0.0089 | 18.0 | 60012 | 0.0271 | 0 | 13 | | 0.0085 | 19.0 | 63346 | 0.0248 | 0 | 13 | | 0.0074 | 20.0 | 66680 | 0.0277 | 0 | 12 | | 0.007 | 21.0 | 70014 | 0.0309 | 0 | 13 | | 0.0066 | 22.0 | 73348 | 0.0306 | 0 | 11 | | 0.0056 | 23.0 | 76682 | 0.0287 | 0 | 10 | | 0.0053 | 24.0 | 80016 | 0.0312 | 0 | 12 | | 0.0049 | 25.0 | 83350 | 0.0276 | 0 | 11 | | 0.0053 | 26.0 | 86684 | 0.0308 | 0 | 10 | | 0.0041 | 27.0 | 90018 | 0.0279 | 0 | 10 | | 0.0041 | 28.0 | 93352 | 0.0292 | 0 | 11 | | 0.0037 | 29.0 | 96686 | 0.0306 | 0 | 11 | | 0.0035 | 30.0 | 100020 | 0.0272 | 0 | 12 | | 0.0032 | 31.0 | 103354 | 0.0255 | 0 | 9 | | 0.0031 | 32.0 | 106688 | 0.0293 | 0 | 10 | | 0.0029 | 33.0 | 110022 | 0.0300 | 0 | 13 | | 0.0026 | 34.0 | 113356 | 0.0305 | 0 | 11 | | 0.0024 | 35.0 | 116690 | 0.0273 | 0 | 9 | | 0.0023 | 36.0 | 120024 | 0.0284 | 0 | 10 | | 0.0022 | 37.0 | 123358 | 0.0313 | 0 | 13 | | 0.002 | 38.0 | 126692 | 0.0341 | 0 | 13 | | 0.0017 | 39.0 | 130026 | 0.0301 | 0 | 13 | | 0.0017 | 40.0 | 133360 | 0.0330 | 0 | 11 | | 0.0016 | 41.0 | 136694 | 0.0344 | 0 | 11 | | 0.0014 | 42.0 | 140028 | 0.0337 | 0 | 10 | | 0.0013 | 43.0 | 143362 | 0.0292 | 0 | 12 | | 0.0012 | 44.0 | 146696 | 0.0339 | 0 | 11 | | 0.0012 | 45.0 | 150030 | 0.0330 | 0 | 11 | | 0.001 | 46.0 | 153364 | 0.0307 | 0 | 11 | | 0.001 | 47.0 | 156698 | 0.0330 | 0 | 10 | | 0.0009 | 48.0 | 160032 | 0.0338 | 0 | 11 | | 0.0009 | 49.0 | 163366 | 0.0288 | 0 | 10 | | 0.0008 | 50.0 | 166700 | 0.0256 | 0 | 10 | | 0.0007 | 51.0 | 170034 | 0.0284 | 0 | 11 | | 0.0006 | 52.0 | 173368 | 0.0342 | 0 | 10 | | 0.0006 | 53.0 | 176702 | 0.0312 | 0 | 10 | | 0.0005 | 54.0 | 180036 | 0.0326 | 0 | 10 | | 0.0005 | 55.0 | 183370 | 0.0304 | 0 | 11 | | 0.0005 | 56.0 | 186704 | 0.0300 | 0 | 11 | | 0.0004 | 57.0 | 190038 | 0.0313 | 0 | 11 | | 0.0003 | 58.0 | 193372 | 0.0321 | 0 | 11 | | 0.0003 | 59.0 | 196706 | 0.0316 | 0 | 10 | | 0.0004 | 60.0 | 200040 | 0.0318 | 0 | 11 | ### Framework versions - Transformers 4.32.1 - Pytorch 2.0.1+cu117 - Datasets 2.14.4 - Tokenizers 0.13.3