Commit
•
1fcc08a
1
Parent(s):
939c62e
End of training
Browse files
README.md
ADDED
|
@@ -0,0 +1,92 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
base_model: distilbert/distilbert-base-uncased
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
datasets:
|
| 8 |
+
- wnut_17
|
| 9 |
+
metrics:
|
| 10 |
+
- precision
|
| 11 |
+
- recall
|
| 12 |
+
- f1
|
| 13 |
+
- accuracy
|
| 14 |
+
model-index:
|
| 15 |
+
- name: distilbert_ner_wnut_model
|
| 16 |
+
results:
|
| 17 |
+
- task:
|
| 18 |
+
name: Token Classification
|
| 19 |
+
type: token-classification
|
| 20 |
+
dataset:
|
| 21 |
+
name: wnut_17
|
| 22 |
+
type: wnut_17
|
| 23 |
+
config: wnut_17
|
| 24 |
+
split: test
|
| 25 |
+
args: wnut_17
|
| 26 |
+
metrics:
|
| 27 |
+
- name: Precision
|
| 28 |
+
type: precision
|
| 29 |
+
value: 0.5339578454332553
|
| 30 |
+
- name: Recall
|
| 31 |
+
type: recall
|
| 32 |
+
value: 0.211306765523633
|
| 33 |
+
- name: F1
|
| 34 |
+
type: f1
|
| 35 |
+
value: 0.30278884462151395
|
| 36 |
+
- name: Accuracy
|
| 37 |
+
type: accuracy
|
| 38 |
+
value: 0.9365995468342525
|
| 39 |
+
---
|
| 40 |
+
|
| 41 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 42 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 43 |
+
|
| 44 |
+
# distilbert_ner_wnut_model
|
| 45 |
+
|
| 46 |
+
This model is a fine-tuned version of [distilbert/distilbert-base-uncased](https://huggingface.co/distilbert/distilbert-base-uncased) on the wnut_17 dataset.
|
| 47 |
+
It achieves the following results on the evaluation set:
|
| 48 |
+
- Loss: 0.2970
|
| 49 |
+
- Precision: 0.5340
|
| 50 |
+
- Recall: 0.2113
|
| 51 |
+
- F1: 0.3028
|
| 52 |
+
- Accuracy: 0.9366
|
| 53 |
+
|
| 54 |
+
## Model description
|
| 55 |
+
|
| 56 |
+
More information needed
|
| 57 |
+
|
| 58 |
+
## Intended uses & limitations
|
| 59 |
+
|
| 60 |
+
More information needed
|
| 61 |
+
|
| 62 |
+
## Training and evaluation data
|
| 63 |
+
|
| 64 |
+
More information needed
|
| 65 |
+
|
| 66 |
+
## Training procedure
|
| 67 |
+
|
| 68 |
+
### Training hyperparameters
|
| 69 |
+
|
| 70 |
+
The following hyperparameters were used during training:
|
| 71 |
+
- learning_rate: 2e-05
|
| 72 |
+
- train_batch_size: 32
|
| 73 |
+
- eval_batch_size: 32
|
| 74 |
+
- seed: 42
|
| 75 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 76 |
+
- lr_scheduler_type: linear
|
| 77 |
+
- num_epochs: 2
|
| 78 |
+
|
| 79 |
+
### Training results
|
| 80 |
+
|
| 81 |
+
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|
| 82 |
+
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
|
| 83 |
+
| No log | 1.0 | 107 | 0.3140 | 0.3744 | 0.0732 | 0.1225 | 0.9306 |
|
| 84 |
+
| No log | 2.0 | 214 | 0.2970 | 0.5340 | 0.2113 | 0.3028 | 0.9366 |
|
| 85 |
+
|
| 86 |
+
|
| 87 |
+
### Framework versions
|
| 88 |
+
|
| 89 |
+
- Transformers 4.45.0.dev0
|
| 90 |
+
- Pytorch 2.2.1+cu118
|
| 91 |
+
- Datasets 2.21.0
|
| 92 |
+
- Tokenizers 0.19.1
|