File size: 4,874 Bytes
55de14d |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 |
---
base_model: cardiffnlp/twitter-roberta-base-sentiment
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
- precision
- recall
model-index:
- name: x-robertta
results: []
---
<!-- 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. -->
# x-robertta
This model is a fine-tuned version of [cardiffnlp/twitter-roberta-base-sentiment](https://huggingface.co/cardiffnlp/twitter-roberta-base-sentiment) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.4082
- Accuracy: 0.8448
- F1: 0.8443
- Precision: 0.8439
- Recall: 0.8452
## 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
- lr_scheduler_warmup_steps: 100
- num_epochs: 3
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|:-------------:|:------:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
| 0.9847 | 0.0820 | 50 | 0.9092 | 0.6111 | 0.4914 | 0.4110 | 0.6129 |
| 0.8446 | 0.1639 | 100 | 1.2047 | 0.4515 | 0.4025 | 0.5378 | 0.4543 |
| 0.794 | 0.2459 | 150 | 0.6341 | 0.7058 | 0.6924 | 0.7562 | 0.7044 |
| 0.6176 | 0.3279 | 200 | 0.5220 | 0.8013 | 0.7958 | 0.7994 | 0.8019 |
| 0.6387 | 0.4098 | 250 | 0.5844 | 0.7790 | 0.7668 | 0.7832 | 0.7799 |
| 0.5845 | 0.4918 | 300 | 0.5524 | 0.7897 | 0.7834 | 0.7895 | 0.7906 |
| 0.5467 | 0.5738 | 350 | 0.5331 | 0.8099 | 0.8088 | 0.8089 | 0.8105 |
| 0.5181 | 0.6557 | 400 | 0.5041 | 0.8144 | 0.8118 | 0.8174 | 0.8143 |
| 0.4963 | 0.7377 | 450 | 0.4705 | 0.8228 | 0.8181 | 0.8219 | 0.8234 |
| 0.4871 | 0.8197 | 500 | 0.5085 | 0.8014 | 0.8004 | 0.8133 | 0.8010 |
| 0.5346 | 0.9016 | 550 | 0.4812 | 0.8298 | 0.8232 | 0.8338 | 0.8304 |
| 0.4424 | 0.9836 | 600 | 0.4802 | 0.8319 | 0.8271 | 0.8334 | 0.8323 |
| 0.4791 | 1.0656 | 650 | 0.4963 | 0.8111 | 0.8117 | 0.8149 | 0.8116 |
| 0.4785 | 1.1475 | 700 | 0.4522 | 0.8283 | 0.8279 | 0.8287 | 0.8284 |
| 0.4196 | 1.2295 | 750 | 0.5025 | 0.8124 | 0.8104 | 0.8183 | 0.8122 |
| 0.4284 | 1.3115 | 800 | 0.4800 | 0.8191 | 0.8189 | 0.8209 | 0.8196 |
| 0.4312 | 1.3934 | 850 | 0.6048 | 0.7608 | 0.7367 | 0.7859 | 0.7621 |
| 0.413 | 1.4754 | 900 | 0.4465 | 0.8412 | 0.8377 | 0.8409 | 0.8416 |
| 0.4239 | 1.5574 | 950 | 0.4960 | 0.8172 | 0.8172 | 0.8211 | 0.8178 |
| 0.4354 | 1.6393 | 1000 | 0.4348 | 0.8325 | 0.8328 | 0.8360 | 0.8324 |
| 0.4172 | 1.7213 | 1050 | 0.4525 | 0.8341 | 0.8298 | 0.8365 | 0.8344 |
| 0.4384 | 1.8033 | 1100 | 0.4169 | 0.8442 | 0.8416 | 0.8445 | 0.8444 |
| 0.4402 | 1.8852 | 1150 | 0.4124 | 0.8430 | 0.8405 | 0.8415 | 0.8433 |
| 0.4232 | 1.9672 | 1200 | 0.4187 | 0.8406 | 0.8388 | 0.8423 | 0.8407 |
| 0.3738 | 2.0492 | 1250 | 0.4367 | 0.8422 | 0.8413 | 0.8434 | 0.8422 |
| 0.373 | 2.1311 | 1300 | 0.4338 | 0.8415 | 0.8407 | 0.8434 | 0.8415 |
| 0.369 | 2.2131 | 1350 | 0.4468 | 0.8385 | 0.8395 | 0.8412 | 0.8387 |
| 0.3772 | 2.2951 | 1400 | 0.4141 | 0.8461 | 0.8452 | 0.8455 | 0.8462 |
| 0.3602 | 2.3770 | 1450 | 0.4495 | 0.8214 | 0.8235 | 0.8359 | 0.8211 |
| 0.3735 | 2.4590 | 1500 | 0.4055 | 0.8456 | 0.8449 | 0.8449 | 0.8458 |
| 0.3585 | 2.5410 | 1550 | 0.4115 | 0.8470 | 0.8450 | 0.8463 | 0.8472 |
| 0.3795 | 2.6230 | 1600 | 0.4318 | 0.8372 | 0.8364 | 0.8368 | 0.8377 |
| 0.356 | 2.7049 | 1650 | 0.4179 | 0.8434 | 0.8440 | 0.8446 | 0.8435 |
| 0.3554 | 2.7869 | 1700 | 0.4080 | 0.8476 | 0.8471 | 0.8473 | 0.8477 |
| 0.3729 | 2.8689 | 1750 | 0.4044 | 0.8491 | 0.8478 | 0.8479 | 0.8494 |
| 0.3578 | 2.9508 | 1800 | 0.4030 | 0.8482 | 0.8473 | 0.8474 | 0.8484 |
### Framework versions
- Transformers 4.42.4
- Pytorch 2.3.1+cu121
- Tokenizers 0.19.1
|