|
--- |
|
license: apache-2.0 |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- wmt16 |
|
model-index: |
|
- name: t5-turkish-to-english |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# t5-turkish-to-english |
|
|
|
This model is a fine-tuned version of [t5-base](https://huggingface.co/t5-base) on the wmt16 dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.0282 |
|
|
|
## 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: 5e-05 |
|
- train_batch_size: 8 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 500 |
|
- num_epochs: 3 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:-----:|:---------------:| |
|
| 0.6168 | 0.02 | 500 | 0.0497 | |
|
| 0.0832 | 0.04 | 1000 | 0.0448 | |
|
| 0.0791 | 0.06 | 1500 | 0.0424 | |
|
| 0.0718 | 0.08 | 2000 | 0.0413 | |
|
| 0.0661 | 0.1 | 2500 | 0.0406 | |
|
| 0.0669 | 0.12 | 3000 | 0.0399 | |
|
| 0.065 | 0.14 | 3500 | 0.0389 | |
|
| 0.0627 | 0.16 | 4000 | 0.0389 | |
|
| 0.0637 | 0.17 | 4500 | 0.0396 | |
|
| 0.0599 | 0.19 | 5000 | 0.0376 | |
|
| 0.0601 | 0.21 | 5500 | 0.0368 | |
|
| 0.0594 | 0.23 | 6000 | 0.0379 | |
|
| 0.0578 | 0.25 | 6500 | 0.0371 | |
|
| 0.0577 | 0.27 | 7000 | 0.0383 | |
|
| 0.0566 | 0.29 | 7500 | 0.0377 | |
|
| 0.0554 | 0.31 | 8000 | 0.0351 | |
|
| 0.0554 | 0.33 | 8500 | 0.0347 | |
|
| 0.0546 | 0.35 | 9000 | 0.0351 | |
|
| 0.0564 | 0.37 | 9500 | 0.0356 | |
|
| 0.0533 | 0.39 | 10000 | 0.0340 | |
|
| 0.0515 | 0.41 | 10500 | 0.0339 | |
|
| 0.0523 | 0.43 | 11000 | 0.0337 | |
|
| 0.0528 | 0.45 | 11500 | 0.0337 | |
|
| 0.0536 | 0.47 | 12000 | 0.0332 | |
|
| 0.0501 | 0.49 | 12500 | 0.0334 | |
|
| 0.0493 | 0.51 | 13000 | 0.0332 | |
|
| 0.0504 | 0.52 | 13500 | 0.0331 | |
|
| 0.0484 | 0.54 | 14000 | 0.0328 | |
|
| 0.0496 | 0.56 | 14500 | 0.0327 | |
|
| 0.0469 | 0.58 | 15000 | 0.0331 | |
|
| 0.0483 | 0.6 | 15500 | 0.0329 | |
|
| 0.0477 | 0.62 | 16000 | 0.0326 | |
|
| 0.0492 | 0.64 | 16500 | 0.0326 | |
|
| 0.0482 | 0.66 | 17000 | 0.0322 | |
|
| 0.0468 | 0.68 | 17500 | 0.0323 | |
|
| 0.0474 | 0.7 | 18000 | 0.0320 | |
|
| 0.0463 | 0.72 | 18500 | 0.0321 | |
|
| 0.048 | 0.74 | 19000 | 0.0319 | |
|
| 0.0463 | 0.76 | 19500 | 0.0319 | |
|
| 0.0467 | 0.78 | 20000 | 0.0316 | |
|
| 0.0457 | 0.8 | 20500 | 0.0319 | |
|
| 0.0463 | 0.82 | 21000 | 0.0320 | |
|
| 0.045 | 0.84 | 21500 | 0.0317 | |
|
| 0.0442 | 0.86 | 22000 | 0.0314 | |
|
| 0.0462 | 0.87 | 22500 | 0.0313 | |
|
| 0.0453 | 0.89 | 23000 | 0.0313 | |
|
| 0.0455 | 0.91 | 23500 | 0.0316 | |
|
| 0.0459 | 0.93 | 24000 | 0.0311 | |
|
| 0.0435 | 0.95 | 24500 | 0.0312 | |
|
| 0.0451 | 0.97 | 25000 | 0.0310 | |
|
| 0.043 | 0.99 | 25500 | 0.0310 | |
|
| 0.0429 | 1.01 | 26000 | 0.0306 | |
|
| 0.0423 | 1.03 | 26500 | 0.0309 | |
|
| 0.0418 | 1.05 | 27000 | 0.0309 | |
|
| 0.0418 | 1.07 | 27500 | 0.0308 | |
|
| 0.0414 | 1.09 | 28000 | 0.0307 | |
|
| 0.0426 | 1.11 | 28500 | 0.0308 | |
|
| 0.0411 | 1.13 | 29000 | 0.0306 | |
|
| 0.0414 | 1.15 | 29500 | 0.0310 | |
|
| 0.0411 | 1.17 | 30000 | 0.0305 | |
|
| 0.0424 | 1.19 | 30500 | 0.0305 | |
|
| 0.0419 | 1.21 | 31000 | 0.0307 | |
|
| 0.0415 | 1.22 | 31500 | 0.0304 | |
|
| 0.0403 | 1.24 | 32000 | 0.0303 | |
|
| 0.0411 | 1.26 | 32500 | 0.0302 | |
|
| 0.0414 | 1.28 | 33000 | 0.0304 | |
|
| 0.0412 | 1.3 | 33500 | 0.0301 | |
|
| 0.0404 | 1.32 | 34000 | 0.0304 | |
|
| 0.0403 | 1.34 | 34500 | 0.0304 | |
|
| 0.0415 | 1.36 | 35000 | 0.0302 | |
|
| 0.0389 | 1.38 | 35500 | 0.0303 | |
|
| 0.0401 | 1.4 | 36000 | 0.0300 | |
|
| 0.0393 | 1.42 | 36500 | 0.0301 | |
|
| 0.0399 | 1.44 | 37000 | 0.0297 | |
|
| 0.0404 | 1.46 | 37500 | 0.0297 | |
|
| 0.0404 | 1.48 | 38000 | 0.0298 | |
|
| 0.04 | 1.5 | 38500 | 0.0296 | |
|
| 0.0403 | 1.52 | 39000 | 0.0296 | |
|
| 0.04 | 1.54 | 39500 | 0.0294 | |
|
| 0.0392 | 1.56 | 40000 | 0.0295 | |
|
| 0.0392 | 1.57 | 40500 | 0.0295 | |
|
| 0.0388 | 1.59 | 41000 | 0.0296 | |
|
| 0.0398 | 1.61 | 41500 | 0.0297 | |
|
| 0.0388 | 1.63 | 42000 | 0.0293 | |
|
| 0.0385 | 1.65 | 42500 | 0.0294 | |
|
| 0.0392 | 1.67 | 43000 | 0.0291 | |
|
| 0.0384 | 1.69 | 43500 | 0.0293 | |
|
| 0.0384 | 1.71 | 44000 | 0.0294 | |
|
| 0.0395 | 1.73 | 44500 | 0.0291 | |
|
| 0.0391 | 1.75 | 45000 | 0.0293 | |
|
| 0.0375 | 1.77 | 45500 | 0.0293 | |
|
| 0.0375 | 1.79 | 46000 | 0.0294 | |
|
| 0.0388 | 1.81 | 46500 | 0.0292 | |
|
| 0.0392 | 1.83 | 47000 | 0.0293 | |
|
| 0.0382 | 1.85 | 47500 | 0.0294 | |
|
| 0.038 | 1.87 | 48000 | 0.0293 | |
|
| 0.0388 | 1.89 | 48500 | 0.0292 | |
|
| 0.0383 | 1.91 | 49000 | 0.0290 | |
|
| 0.0381 | 1.92 | 49500 | 0.0292 | |
|
| 0.0388 | 1.94 | 50000 | 0.0290 | |
|
| 0.0378 | 1.96 | 50500 | 0.0289 | |
|
| 0.0391 | 1.98 | 51000 | 0.0290 | |
|
| 0.0379 | 2.0 | 51500 | 0.0289 | |
|
| 0.0364 | 2.02 | 52000 | 0.0289 | |
|
| 0.0366 | 2.04 | 52500 | 0.0291 | |
|
| 0.0362 | 2.06 | 53000 | 0.0291 | |
|
| 0.0359 | 2.08 | 53500 | 0.0289 | |
|
| 0.0367 | 2.1 | 54000 | 0.0291 | |
|
| 0.0368 | 2.12 | 54500 | 0.0290 | |
|
| 0.0359 | 2.14 | 55000 | 0.0288 | |
|
| 0.0359 | 2.16 | 55500 | 0.0289 | |
|
| 0.036 | 2.18 | 56000 | 0.0289 | |
|
| 0.0362 | 2.2 | 56500 | 0.0288 | |
|
| 0.0359 | 2.22 | 57000 | 0.0287 | |
|
| 0.0374 | 2.24 | 57500 | 0.0287 | |
|
| 0.0353 | 2.26 | 58000 | 0.0286 | |
|
| 0.0351 | 2.27 | 58500 | 0.0287 | |
|
| 0.0348 | 2.29 | 59000 | 0.0286 | |
|
| 0.0355 | 2.31 | 59500 | 0.0286 | |
|
| 0.0362 | 2.33 | 60000 | 0.0287 | |
|
| 0.0361 | 2.35 | 60500 | 0.0287 | |
|
| 0.0354 | 2.37 | 61000 | 0.0286 | |
|
| 0.036 | 2.39 | 61500 | 0.0284 | |
|
| 0.0341 | 2.41 | 62000 | 0.0285 | |
|
| 0.0348 | 2.43 | 62500 | 0.0284 | |
|
| 0.036 | 2.45 | 63000 | 0.0285 | |
|
| 0.0351 | 2.47 | 63500 | 0.0284 | |
|
| 0.0354 | 2.49 | 64000 | 0.0284 | |
|
| 0.0372 | 2.51 | 64500 | 0.0285 | |
|
| 0.035 | 2.53 | 65000 | 0.0285 | |
|
| 0.0348 | 2.55 | 65500 | 0.0284 | |
|
| 0.0353 | 2.57 | 66000 | 0.0283 | |
|
| 0.0353 | 2.59 | 66500 | 0.0283 | |
|
| 0.0352 | 2.6 | 67000 | 0.0283 | |
|
| 0.0357 | 2.62 | 67500 | 0.0283 | |
|
| 0.035 | 2.64 | 68000 | 0.0283 | |
|
| 0.0352 | 2.66 | 68500 | 0.0283 | |
|
| 0.035 | 2.68 | 69000 | 0.0282 | |
|
| 0.0348 | 2.7 | 69500 | 0.0282 | |
|
| 0.0344 | 2.72 | 70000 | 0.0281 | |
|
| 0.0357 | 2.74 | 70500 | 0.0282 | |
|
| 0.0348 | 2.76 | 71000 | 0.0282 | |
|
| 0.0349 | 2.78 | 71500 | 0.0281 | |
|
| 0.0365 | 2.8 | 72000 | 0.0282 | |
|
| 0.0354 | 2.82 | 72500 | 0.0282 | |
|
| 0.0359 | 2.84 | 73000 | 0.0281 | |
|
| 0.0343 | 2.86 | 73500 | 0.0282 | |
|
| 0.0343 | 2.88 | 74000 | 0.0281 | |
|
| 0.0346 | 2.9 | 74500 | 0.0282 | |
|
| 0.0357 | 2.92 | 75000 | 0.0282 | |
|
| 0.0351 | 2.94 | 75500 | 0.0282 | |
|
| 0.0355 | 2.95 | 76000 | 0.0282 | |
|
| 0.0351 | 2.97 | 76500 | 0.0282 | |
|
| 0.0359 | 2.99 | 77000 | 0.0282 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.20.1 |
|
- Pytorch 1.11.0 |
|
- Datasets 2.1.0 |
|
- Tokenizers 0.12.1 |
|
|