End of training
Browse files
README.md
ADDED
|
@@ -0,0 +1,83 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: casual/nlp_til
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
metrics:
|
| 7 |
+
- precision
|
| 8 |
+
- recall
|
| 9 |
+
- f1
|
| 10 |
+
- accuracy
|
| 11 |
+
model-index:
|
| 12 |
+
- name: nlp_til2
|
| 13 |
+
results: []
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 17 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 18 |
+
|
| 19 |
+
# nlp_til2
|
| 20 |
+
|
| 21 |
+
This model is a fine-tuned version of [casual/nlp_til](https://huggingface.co/casual/nlp_til) on an unknown dataset.
|
| 22 |
+
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 0.0989
|
| 24 |
+
- Precision: 0.7550
|
| 25 |
+
- Recall: 0.7387
|
| 26 |
+
- F1: 0.7468
|
| 27 |
+
- Accuracy: 0.9573
|
| 28 |
+
|
| 29 |
+
## Model description
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Intended uses & limitations
|
| 34 |
+
|
| 35 |
+
More information needed
|
| 36 |
+
|
| 37 |
+
## Training and evaluation data
|
| 38 |
+
|
| 39 |
+
More information needed
|
| 40 |
+
|
| 41 |
+
## Training procedure
|
| 42 |
+
|
| 43 |
+
### Training hyperparameters
|
| 44 |
+
|
| 45 |
+
The following hyperparameters were used during training:
|
| 46 |
+
- learning_rate: 2e-05
|
| 47 |
+
- train_batch_size: 16
|
| 48 |
+
- eval_batch_size: 16
|
| 49 |
+
- seed: 42
|
| 50 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 51 |
+
- lr_scheduler_type: linear
|
| 52 |
+
- num_epochs: 18
|
| 53 |
+
|
| 54 |
+
### Training results
|
| 55 |
+
|
| 56 |
+
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|
| 57 |
+
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
|
| 58 |
+
| No log | 1.0 | 219 | 0.2111 | 0.4478 | 0.5230 | 0.4825 | 0.8922 |
|
| 59 |
+
| No log | 2.0 | 438 | 0.1953 | 0.4984 | 0.5128 | 0.5055 | 0.9040 |
|
| 60 |
+
| 0.2163 | 3.0 | 657 | 0.1886 | 0.5210 | 0.5660 | 0.5426 | 0.9095 |
|
| 61 |
+
| 0.2163 | 4.0 | 876 | 0.1809 | 0.5432 | 0.5988 | 0.5696 | 0.9142 |
|
| 62 |
+
| 0.2056 | 5.0 | 1095 | 0.1692 | 0.5758 | 0.6142 | 0.5944 | 0.9207 |
|
| 63 |
+
| 0.2056 | 6.0 | 1314 | 0.1625 | 0.5889 | 0.6232 | 0.6056 | 0.9247 |
|
| 64 |
+
| 0.188 | 7.0 | 1533 | 0.1510 | 0.6315 | 0.5979 | 0.6142 | 0.9317 |
|
| 65 |
+
| 0.188 | 8.0 | 1752 | 0.1405 | 0.6625 | 0.6341 | 0.6480 | 0.9373 |
|
| 66 |
+
| 0.188 | 9.0 | 1971 | 0.1341 | 0.6665 | 0.6576 | 0.6620 | 0.9399 |
|
| 67 |
+
| 0.1716 | 10.0 | 2190 | 0.1305 | 0.6594 | 0.6954 | 0.6769 | 0.9409 |
|
| 68 |
+
| 0.1716 | 11.0 | 2409 | 0.1221 | 0.6931 | 0.6897 | 0.6914 | 0.9455 |
|
| 69 |
+
| 0.1565 | 12.0 | 2628 | 0.1185 | 0.6970 | 0.7239 | 0.7102 | 0.9477 |
|
| 70 |
+
| 0.1565 | 13.0 | 2847 | 0.1082 | 0.7336 | 0.7087 | 0.7210 | 0.9523 |
|
| 71 |
+
| 0.1459 | 14.0 | 3066 | 0.1079 | 0.7323 | 0.7302 | 0.7312 | 0.9532 |
|
| 72 |
+
| 0.1459 | 15.0 | 3285 | 0.1039 | 0.7325 | 0.7251 | 0.7287 | 0.9542 |
|
| 73 |
+
| 0.1419 | 16.0 | 3504 | 0.1030 | 0.7389 | 0.7484 | 0.7436 | 0.9557 |
|
| 74 |
+
| 0.1419 | 17.0 | 3723 | 0.1001 | 0.7487 | 0.7414 | 0.7450 | 0.9565 |
|
| 75 |
+
| 0.1419 | 18.0 | 3942 | 0.0989 | 0.7550 | 0.7387 | 0.7468 | 0.9573 |
|
| 76 |
+
|
| 77 |
+
|
| 78 |
+
### Framework versions
|
| 79 |
+
|
| 80 |
+
- Transformers 4.40.2
|
| 81 |
+
- Pytorch 2.0.1+cu117
|
| 82 |
+
- Datasets 2.19.1
|
| 83 |
+
- Tokenizers 0.19.1
|