trainer_3f
This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.0618
- Precision: 0.8364
- Recall: 0.8325
- F1: 0.8329
- Accuracy: 0.8325
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
0.0336 | 0.08 | 30 | 0.8889 | 0.8378 | 0.8272 | 0.8274 | 0.8272 |
0.0653 | 0.17 | 60 | 0.8057 | 0.8477 | 0.8448 | 0.8446 | 0.8448 |
0.0424 | 0.25 | 90 | 0.8184 | 0.8437 | 0.8430 | 0.8430 | 0.8430 |
0.0725 | 0.34 | 120 | 0.8139 | 0.8490 | 0.8430 | 0.8432 | 0.8430 |
0.0887 | 0.42 | 150 | 0.8627 | 0.8411 | 0.8342 | 0.8341 | 0.8342 |
0.0441 | 0.51 | 180 | 0.8865 | 0.8505 | 0.8413 | 0.8420 | 0.8413 |
0.0085 | 0.59 | 210 | 0.8662 | 0.8458 | 0.8430 | 0.8428 | 0.8430 |
0.0785 | 0.67 | 240 | 0.9317 | 0.8431 | 0.8395 | 0.8398 | 0.8395 |
0.0075 | 0.76 | 270 | 0.9654 | 0.8538 | 0.8448 | 0.8446 | 0.8448 |
0.0259 | 0.84 | 300 | 0.9987 | 0.8405 | 0.8342 | 0.8342 | 0.8342 |
0.0672 | 0.93 | 330 | 1.0019 | 0.8394 | 0.8342 | 0.8346 | 0.8342 |
0.017 | 1.01 | 360 | 1.0248 | 0.8384 | 0.8325 | 0.8332 | 0.8325 |
0.0049 | 1.1 | 390 | 1.0067 | 0.8474 | 0.8377 | 0.8393 | 0.8377 |
0.0296 | 1.18 | 420 | 1.0234 | 0.8438 | 0.8377 | 0.8385 | 0.8377 |
0.0349 | 1.26 | 450 | 1.0185 | 0.8277 | 0.8219 | 0.8223 | 0.8219 |
0.0025 | 1.35 | 480 | 1.0467 | 0.8239 | 0.8166 | 0.8171 | 0.8166 |
0.0041 | 1.43 | 510 | 1.0966 | 0.8315 | 0.8236 | 0.8243 | 0.8236 |
0.0017 | 1.52 | 540 | 1.0549 | 0.8323 | 0.8272 | 0.8279 | 0.8272 |
0.0361 | 1.6 | 570 | 1.0055 | 0.8519 | 0.8483 | 0.8490 | 0.8483 |
0.0644 | 1.69 | 600 | 1.1315 | 0.8371 | 0.8325 | 0.8322 | 0.8325 |
0.0016 | 1.77 | 630 | 1.1434 | 0.8244 | 0.8183 | 0.8188 | 0.8183 |
0.0405 | 1.85 | 660 | 1.0628 | 0.8326 | 0.8272 | 0.8276 | 0.8272 |
0.0005 | 1.94 | 690 | 1.0394 | 0.8391 | 0.8342 | 0.8348 | 0.8342 |
0.0324 | 2.02 | 720 | 1.1081 | 0.8316 | 0.8254 | 0.8264 | 0.8254 |
0.0012 | 2.11 | 750 | 1.0663 | 0.8354 | 0.8325 | 0.8323 | 0.8325 |
0.005 | 2.19 | 780 | 1.0777 | 0.8335 | 0.8307 | 0.8304 | 0.8307 |
0.0264 | 2.28 | 810 | 1.0483 | 0.8361 | 0.8325 | 0.8326 | 0.8325 |
0.0431 | 2.36 | 840 | 1.0193 | 0.8473 | 0.8430 | 0.8435 | 0.8430 |
0.0004 | 2.44 | 870 | 1.0411 | 0.8457 | 0.8413 | 0.8418 | 0.8413 |
0.0045 | 2.53 | 900 | 1.0604 | 0.8319 | 0.8289 | 0.8290 | 0.8289 |
0.0305 | 2.61 | 930 | 1.0808 | 0.8323 | 0.8289 | 0.8292 | 0.8289 |
0.0362 | 2.7 | 960 | 1.0466 | 0.8430 | 0.8395 | 0.8398 | 0.8395 |
0.0004 | 2.78 | 990 | 1.0518 | 0.8429 | 0.8395 | 0.8397 | 0.8395 |
0.0147 | 2.87 | 1020 | 1.0781 | 0.8397 | 0.8360 | 0.8361 | 0.8360 |
0.0034 | 2.95 | 1050 | 1.0696 | 0.8377 | 0.8342 | 0.8344 | 0.8342 |
0.0004 | 3.03 | 1080 | 1.0649 | 0.8395 | 0.8360 | 0.8362 | 0.8360 |
0.0063 | 3.12 | 1110 | 1.0614 | 0.8347 | 0.8325 | 0.8325 | 0.8325 |
0.0014 | 3.2 | 1140 | 1.0433 | 0.8367 | 0.8342 | 0.8345 | 0.8342 |
0.0181 | 3.29 | 1170 | 1.0559 | 0.8339 | 0.8307 | 0.8311 | 0.8307 |
0.0069 | 3.37 | 1200 | 1.0693 | 0.8320 | 0.8289 | 0.8293 | 0.8289 |
0.0004 | 3.46 | 1230 | 1.0666 | 0.8320 | 0.8289 | 0.8293 | 0.8289 |
0.0117 | 3.54 | 1260 | 1.0596 | 0.8316 | 0.8289 | 0.8292 | 0.8289 |
0.0211 | 3.62 | 1290 | 1.0679 | 0.8326 | 0.8289 | 0.8293 | 0.8289 |
0.0124 | 3.71 | 1320 | 1.0688 | 0.8359 | 0.8325 | 0.8328 | 0.8325 |
0.0177 | 3.79 | 1350 | 1.0676 | 0.8346 | 0.8307 | 0.8312 | 0.8307 |
0.0009 | 3.88 | 1380 | 1.0635 | 0.8364 | 0.8325 | 0.8329 | 0.8325 |
0.0012 | 3.96 | 1410 | 1.0616 | 0.8364 | 0.8325 | 0.8329 | 0.8325 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 119
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
Model tree for SimoneJLaudani/trainer_3f
Base model
distilbert/distilbert-base-uncased