<!-- 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. -->
bert-base-uncased-Twitter_Sentiment_Analysis_v2
This model is a fine-tuned version of bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5809
- Accuracy: 0.8522
- Weighted f1: 0.8507
- Micro f1: 0.8522
- Macro f1: 0.8007
- Weighted recall: 0.8522
- Micro recall: 0.8522
- Macro recall: 0.8006
- Weighted precision: 0.8503
- Micro precision: 0.8522
- Macro precision: 0.8025
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: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 12
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | Weighted f1 | Micro f1 | Macro f1 | Weighted recall | Micro recall | Macro recall | Weighted precision | Micro precision | Macro precision |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
0.3896 | 1.0 | 183 | 0.3073 | 0.8022 | 0.7743 | 0.8022 | 0.6960 | 0.8022 | 0.8022 | 0.6914 | 0.8053 | 0.8022 | 0.7880 |
0.2488 | 2.0 | 366 | 0.2937 | 0.8474 | 0.8409 | 0.8474 | 0.7880 | 0.8474 | 0.8474 | 0.7747 | 0.8429 | 0.8474 | 0.8132 |
0.1766 | 3.0 | 549 | 0.3115 | 0.8398 | 0.8298 | 0.8398 | 0.7750 | 0.8398 | 0.8398 | 0.7613 | 0.8355 | 0.8398 | 0.8111 |
0.1253 | 4.0 | 732 | 0.3354 | 0.8487 | 0.8406 | 0.8487 | 0.7843 | 0.8487 | 0.8487 | 0.7695 | 0.8448 | 0.8487 | 0.8167 |
0.09 | 5.0 | 915 | 0.4137 | 0.8474 | 0.8447 | 0.8474 | 0.7930 | 0.8474 | 0.8474 | 0.7932 | 0.8455 | 0.8474 | 0.7985 |
0.0677 | 6.0 | 1098 | 0.4872 | 0.8494 | 0.8491 | 0.8494 | 0.7977 | 0.8494 | 0.8494 | 0.8068 | 0.8511 | 0.8494 | 0.7930 |
0.0501 | 7.0 | 1281 | 0.4959 | 0.8576 | 0.8556 | 0.8576 | 0.8066 | 0.8576 | 0.8576 | 0.8063 | 0.8558 | 0.8576 | 0.8105 |
0.0415 | 8.0 | 1464 | 0.5412 | 0.8515 | 0.8505 | 0.8515 | 0.8003 | 0.8515 | 0.8515 | 0.8017 | 0.8503 | 0.8515 | 0.8000 |
0.0323 | 9.0 | 1647 | 0.5969 | 0.8480 | 0.8480 | 0.8480 | 0.7939 | 0.8480 | 0.8480 | 0.7918 | 0.8481 | 0.8480 | 0.7961 |
0.0253 | 10.0 | 1830 | 0.5560 | 0.8549 | 0.8526 | 0.8549 | 0.8024 | 0.8549 | 0.8549 | 0.8016 | 0.8527 | 0.8549 | 0.8073 |
0.0204 | 11.0 | 2013 | 0.5697 | 0.8522 | 0.8494 | 0.8522 | 0.7970 | 0.8522 | 0.8522 | 0.7840 | 0.8482 | 0.8522 | 0.8122 |
0.019 | 12.0 | 2196 | 0.5809 | 0.8522 | 0.8507 | 0.8522 | 0.8007 | 0.8522 | 0.8522 | 0.8006 | 0.8503 | 0.8522 | 0.8025 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.12.1
- Datasets 2.9.0
- Tokenizers 0.12.1