<!-- 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. -->
mini-mlm-imdb
This model is a fine-tuned version of google/bert_uncased_L-4_H-256_A-4 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.7643
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: 3e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- num_epochs: 200
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
3.2543 | 0.16 | 500 | 3.0042 |
3.1664 | 0.32 | 1000 | 2.9716 |
3.1428 | 0.48 | 1500 | 2.9460 |
3.1363 | 0.64 | 2000 | 2.9316 |
3.1088 | 0.8 | 2500 | 2.9068 |
3.0943 | 0.96 | 3000 | 2.9046 |
3.0436 | 1.12 | 3500 | 2.8918 |
3.0326 | 1.28 | 4000 | 2.8762 |
3.0302 | 1.44 | 4500 | 2.8765 |
3.0232 | 1.6 | 5000 | 2.8658 |
3.0123 | 1.76 | 5500 | 2.8538 |
3.0164 | 1.92 | 6000 | 2.8530 |
2.992 | 2.08 | 6500 | 2.8487 |
2.9922 | 2.24 | 7000 | 2.8440 |
2.9862 | 2.4 | 7500 | 2.8348 |
2.9621 | 2.56 | 8000 | 2.8324 |
2.9926 | 2.72 | 8500 | 2.8235 |
2.9871 | 2.88 | 9000 | 2.8223 |
2.9593 | 3.04 | 9500 | 2.8131 |
2.9404 | 3.2 | 10000 | 2.8119 |
2.9278 | 3.36 | 10500 | 2.8076 |
2.943 | 3.52 | 11000 | 2.8015 |
2.9074 | 3.68 | 11500 | 2.8067 |
2.9247 | 3.84 | 12000 | 2.8027 |
2.9188 | 4.0 | 12500 | 2.7975 |
2.9011 | 4.16 | 13000 | 2.7905 |
2.8973 | 4.32 | 13500 | 2.7893 |
2.8796 | 4.48 | 14000 | 2.7915 |
2.9026 | 4.64 | 14500 | 2.7787 |
2.9022 | 4.8 | 15000 | 2.7819 |
2.8942 | 4.96 | 15500 | 2.7843 |
2.8844 | 5.12 | 16000 | 2.7771 |
2.8777 | 5.28 | 16500 | 2.7701 |
2.8899 | 5.44 | 17000 | 2.7778 |
2.8973 | 5.6 | 17500 | 2.7702 |
2.877 | 5.76 | 18000 | 2.7592 |
2.8704 | 5.92 | 18500 | 2.7711 |
2.8649 | 6.08 | 19000 | 2.7610 |
2.8619 | 6.24 | 19500 | 2.7643 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.12.1
- Datasets 2.7.1
- Tokenizers 0.13.2