<!-- 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. -->
NoDuplicates
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: 0.4279
- Accuracy: 0.9128
- F1 Macro: 0.8384
- F1 Class 0: 0.9406
- F1 Class 1: 0.3333
- F1 Class 2: 0.9127
- F1 Class 3: 0.6471
- F1 Class 4: 0.8254
- F1 Class 5: 0.8293
- F1 Class 6: 0.8767
- F1 Class 7: 0.7606
- F1 Class 8: 0.7500
- F1 Class 9: 0.9878
- F1 Class 10: 0.9444
- F1 Class 11: 0.9630
- F1 Class 12: 0.9265
- F1 Class 13: 0.8980
- F1 Class 14: 0.8444
- F1 Class 15: 0.8132
- F1 Class 16: 0.7778
- F1 Class 17: 0.9651
- F1 Class 18: 0.9574
- F1 Class 19: 0.8148
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
- num_epochs: 3.0
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | F1 Class 0 | F1 Class 1 | F1 Class 2 | F1 Class 3 | F1 Class 4 | F1 Class 5 | F1 Class 6 | F1 Class 7 | F1 Class 8 | F1 Class 9 | F1 Class 10 | F1 Class 11 | F1 Class 12 | F1 Class 13 | F1 Class 14 | F1 Class 15 | F1 Class 16 | F1 Class 17 | F1 Class 18 | F1 Class 19 |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
1.4862 | 0.27 | 300 | 0.8201 | 0.7845 | 0.4484 | 0.8675 | 0.0 | 0.8627 | 0.0 | 0.6733 | 0.0 | 0.6627 | 0.0 | 0.0 | 0.9862 | 0.1935 | 0.9600 | 0.8299 | 0.0833 | 0.2353 | 0.24 | 0.0400 | 0.8852 | 0.9451 | 0.5033 |
0.7269 | 0.53 | 600 | 0.5951 | 0.8491 | 0.6504 | 0.9048 | 0.0 | 0.8567 | 0.0 | 0.7596 | 0.6111 | 0.6887 | 0.0 | 0.0 | 0.9877 | 0.8033 | 0.9286 | 0.8798 | 0.9167 | 0.74 | 0.6857 | 0.5823 | 0.9506 | 0.9485 | 0.7640 |
0.5429 | 0.8 | 900 | 0.5375 | 0.8637 | 0.7086 | 0.8904 | 0.0 | 0.8589 | 0.0 | 0.7254 | 0.7805 | 0.8215 | 0.6769 | 0.0 | 0.9877 | 0.7833 | 1.0 | 0.9022 | 0.9130 | 0.7912 | 0.7733 | 0.7048 | 0.9032 | 0.9474 | 0.7119 |
0.4594 | 1.06 | 1200 | 0.5110 | 0.8805 | 0.7113 | 0.9099 | 0.0 | 0.8925 | 0.0 | 0.7706 | 0.7391 | 0.8139 | 0.4091 | 0.0 | 0.9908 | 0.8785 | 1.0 | 0.8983 | 0.8936 | 0.8090 | 0.7556 | 0.7907 | 0.9529 | 0.9574 | 0.7647 |
0.3484 | 1.33 | 1500 | 0.4679 | 0.8951 | 0.7667 | 0.9180 | 0.0 | 0.9080 | 0.6957 | 0.8 | 0.7619 | 0.8299 | 0.6875 | 0.0 | 0.9908 | 0.8909 | 1.0 | 0.9196 | 0.9130 | 0.8172 | 0.7865 | 0.7527 | 0.9398 | 0.9474 | 0.7755 |
0.3744 | 1.59 | 1800 | 0.4359 | 0.8951 | 0.7774 | 0.9290 | 0.0 | 0.8815 | 0.8462 | 0.8049 | 0.7805 | 0.8449 | 0.7059 | 0.0 | 0.9908 | 0.9346 | 1.0 | 0.9143 | 0.8980 | 0.8387 | 0.7475 | 0.7179 | 0.9647 | 0.9583 | 0.7895 |
0.3514 | 1.86 | 2100 | 0.5161 | 0.8903 | 0.7592 | 0.9109 | 0.0 | 0.8973 | 0.6429 | 0.7603 | 0.7907 | 0.8571 | 0.7077 | 0.0 | 0.9908 | 0.9346 | 1.0 | 0.8971 | 0.8936 | 0.7042 | 0.7324 | 0.7857 | 0.9595 | 0.9574 | 0.7609 |
0.3111 | 2.12 | 2400 | 0.4327 | 0.9080 | 0.8027 | 0.9283 | 0.3333 | 0.9141 | 0.7407 | 0.8207 | 0.8095 | 0.8622 | 0.7606 | 0.0 | 0.9908 | 0.9298 | 0.9630 | 0.9215 | 0.9167 | 0.8041 | 0.8 | 0.8132 | 0.9651 | 0.9574 | 0.8224 |
0.2088 | 2.39 | 2700 | 0.4356 | 0.9128 | 0.8452 | 0.9386 | 0.3333 | 0.9058 | 0.8462 | 0.8265 | 0.8 | 0.8562 | 0.7429 | 0.7500 | 0.9893 | 0.9346 | 0.9630 | 0.9322 | 0.8936 | 0.8205 | 0.8372 | 0.7765 | 0.9651 | 0.9574 | 0.8350 |
0.2317 | 2.65 | 3000 | 0.4294 | 0.9137 | 0.8217 | 0.9365 | 0.3333 | 0.9102 | 0.625 | 0.8243 | 0.8293 | 0.875 | 0.8056 | 0.3333 | 0.9893 | 0.9444 | 0.9630 | 0.9284 | 0.8980 | 0.8478 | 0.8471 | 0.7816 | 0.9651 | 0.9574 | 0.8400 |
0.1816 | 2.92 | 3300 | 0.4279 | 0.9128 | 0.8384 | 0.9406 | 0.3333 | 0.9127 | 0.6471 | 0.8254 | 0.8293 | 0.8767 | 0.7606 | 0.7500 | 0.9878 | 0.9444 | 0.9630 | 0.9265 | 0.8980 | 0.8444 | 0.8132 | 0.7778 | 0.9651 | 0.9574 | 0.8148 |
Framework versions
- Transformers 4.32.0
- Pytorch 2.0.1+cu117
- Datasets 2.14.4
- Tokenizers 0.13.3