Sentiment140_roBERTa_5E
This model is a fine-tuned version of roberta-base on the sentiment140 dataset. It achieves the following results on the evaluation set:
- Loss: 0.4796
- Accuracy: 0.8933
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: 1e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
0.699 | 0.08 | 50 | 0.6734 | 0.5467 |
0.6099 | 0.16 | 100 | 0.4322 | 0.8 |
0.4906 | 0.24 | 150 | 0.3861 | 0.84 |
0.4652 | 0.32 | 200 | 0.4288 | 0.7933 |
0.4874 | 0.4 | 250 | 0.3872 | 0.84 |
0.4735 | 0.48 | 300 | 0.3401 | 0.8667 |
0.3909 | 0.56 | 350 | 0.3484 | 0.84 |
0.4277 | 0.64 | 400 | 0.3207 | 0.88 |
0.3894 | 0.72 | 450 | 0.3310 | 0.8733 |
0.4523 | 0.8 | 500 | 0.3389 | 0.8667 |
0.4087 | 0.88 | 550 | 0.3515 | 0.8467 |
0.3973 | 0.96 | 600 | 0.3513 | 0.8467 |
0.4016 | 1.04 | 650 | 0.3501 | 0.8667 |
0.3613 | 1.12 | 700 | 0.3327 | 0.8667 |
0.343 | 1.2 | 750 | 0.3518 | 0.86 |
0.314 | 1.28 | 800 | 0.3555 | 0.88 |
0.3407 | 1.36 | 850 | 0.3849 | 0.86 |
0.2944 | 1.44 | 900 | 0.3576 | 0.8667 |
0.3267 | 1.52 | 950 | 0.3461 | 0.8733 |
0.3251 | 1.6 | 1000 | 0.3411 | 0.8667 |
0.321 | 1.68 | 1050 | 0.3371 | 0.88 |
0.3057 | 1.76 | 1100 | 0.3322 | 0.88 |
0.3335 | 1.84 | 1150 | 0.3106 | 0.8667 |
0.3363 | 1.92 | 1200 | 0.3158 | 0.8933 |
0.2972 | 2.0 | 1250 | 0.3122 | 0.88 |
0.2453 | 2.08 | 1300 | 0.3327 | 0.8867 |
0.2467 | 2.16 | 1350 | 0.3767 | 0.8667 |
0.273 | 2.24 | 1400 | 0.3549 | 0.8667 |
0.2672 | 2.32 | 1450 | 0.3470 | 0.88 |
0.2352 | 2.4 | 1500 | 0.4092 | 0.8667 |
0.2763 | 2.48 | 1550 | 0.3472 | 0.9 |
0.2858 | 2.56 | 1600 | 0.3440 | 0.9 |
0.2206 | 2.64 | 1650 | 0.3770 | 0.88 |
0.2928 | 2.72 | 1700 | 0.3280 | 0.8867 |
0.2478 | 2.8 | 1750 | 0.3426 | 0.8867 |
0.2362 | 2.88 | 1800 | 0.3578 | 0.8933 |
0.2107 | 2.96 | 1850 | 0.3986 | 0.8933 |
0.2191 | 3.04 | 1900 | 0.3819 | 0.8933 |
0.2267 | 3.12 | 1950 | 0.4047 | 0.8867 |
0.2076 | 3.2 | 2000 | 0.4303 | 0.8867 |
0.1868 | 3.28 | 2050 | 0.4385 | 0.8933 |
0.2239 | 3.36 | 2100 | 0.4175 | 0.8933 |
0.2082 | 3.44 | 2150 | 0.4142 | 0.8933 |
0.2423 | 3.52 | 2200 | 0.4002 | 0.8867 |
0.1878 | 3.6 | 2250 | 0.4662 | 0.88 |
0.1892 | 3.68 | 2300 | 0.4783 | 0.88 |
0.2259 | 3.76 | 2350 | 0.4487 | 0.88 |
0.1859 | 3.84 | 2400 | 0.4456 | 0.8933 |
0.2042 | 3.92 | 2450 | 0.4468 | 0.8933 |
0.2096 | 4.0 | 2500 | 0.4153 | 0.8867 |
0.178 | 4.08 | 2550 | 0.4100 | 0.8933 |
0.1621 | 4.16 | 2600 | 0.4292 | 0.8933 |
0.1682 | 4.24 | 2650 | 0.4602 | 0.8933 |
0.1813 | 4.32 | 2700 | 0.4680 | 0.8933 |
0.2033 | 4.4 | 2750 | 0.4735 | 0.8933 |
0.1662 | 4.48 | 2800 | 0.4750 | 0.88 |
0.1686 | 4.56 | 2850 | 0.4830 | 0.8933 |
0.1603 | 4.64 | 2900 | 0.4909 | 0.8933 |
0.148 | 4.72 | 2950 | 0.4784 | 0.8933 |
0.162 | 4.8 | 3000 | 0.4750 | 0.8867 |
0.153 | 4.88 | 3050 | 0.4759 | 0.8867 |
0.1657 | 4.96 | 3100 | 0.4796 | 0.8933 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.13.0
- Datasets 2.3.2
- Tokenizers 0.13.1
- Downloads last month
- 257
This model does not have enough activity to be deployed to Inference API (serverless) yet.
Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.