<!-- 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. -->
scenario-non-kd-from-pre-finetune-div-2-data-indolem_sentiment-model-xlm-roberta
This model is a fine-tuned version of xlm-roberta-base on the indolem_sentiment dataset. It achieves the following results on the evaluation set:
- Loss: 0.8749
- Accuracy: 0.8471
- F1: 0.7570
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: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 6969
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
---|---|---|---|---|---|
No log | 0.88 | 100 | 0.5662 | 0.6316 | 0.6059 |
No log | 1.75 | 200 | 0.5076 | 0.7719 | 0.5027 |
No log | 2.63 | 300 | 0.4662 | 0.7870 | 0.6840 |
No log | 3.51 | 400 | 0.4502 | 0.8045 | 0.6355 |
0.3694 | 4.39 | 500 | 0.4589 | 0.8271 | 0.7229 |
0.3694 | 5.26 | 600 | 0.6513 | 0.7995 | 0.7122 |
0.3694 | 6.14 | 700 | 0.6380 | 0.8446 | 0.7281 |
0.3694 | 7.02 | 800 | 0.4913 | 0.8396 | 0.7168 |
0.3694 | 7.89 | 900 | 0.5121 | 0.8571 | 0.7554 |
0.1278 | 8.77 | 1000 | 0.6421 | 0.8471 | 0.7589 |
0.1278 | 9.65 | 1100 | 0.8509 | 0.8145 | 0.7357 |
0.1278 | 10.53 | 1200 | 0.6999 | 0.8145 | 0.7319 |
0.1278 | 11.4 | 1300 | 0.7779 | 0.8596 | 0.7607 |
0.1278 | 12.28 | 1400 | 0.6500 | 0.8647 | 0.7672 |
0.0596 | 13.16 | 1500 | 1.0323 | 0.8195 | 0.6129 |
0.0596 | 14.04 | 1600 | 0.7717 | 0.8747 | 0.7863 |
0.0596 | 14.91 | 1700 | 1.1000 | 0.8521 | 0.7378 |
0.0596 | 15.79 | 1800 | 0.8132 | 0.8521 | 0.7592 |
0.0596 | 16.67 | 1900 | 1.1240 | 0.8471 | 0.7265 |
0.0262 | 17.54 | 2000 | 0.9593 | 0.7970 | 0.7138 |
0.0262 | 18.42 | 2100 | 1.0002 | 0.8271 | 0.7435 |
0.0262 | 19.3 | 2200 | 0.8668 | 0.8371 | 0.7451 |
0.0262 | 20.18 | 2300 | 1.2142 | 0.8321 | 0.7331 |
0.0262 | 21.05 | 2400 | 0.9675 | 0.8546 | 0.7434 |
0.0276 | 21.93 | 2500 | 0.8832 | 0.8396 | 0.7377 |
0.0276 | 22.81 | 2600 | 1.0502 | 0.8371 | 0.7280 |
0.0276 | 23.68 | 2700 | 1.0822 | 0.8421 | 0.7407 |
0.0276 | 24.56 | 2800 | 0.8621 | 0.8321 | 0.7287 |
0.0276 | 25.44 | 2900 | 1.2150 | 0.8296 | 0.6881 |
0.0244 | 26.32 | 3000 | 0.9965 | 0.8446 | 0.7257 |
0.0244 | 27.19 | 3100 | 0.8749 | 0.8471 | 0.7570 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.0.1
- Datasets 2.14.5
- Tokenizers 0.13.3