Jorge Lopez Grisman
commited on
Commit
·
bb47ca3
1
Parent(s):
e24d61e
adding readme manually.
Browse files
README.md
ADDED
@@ -0,0 +1,52 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
|
3 |
+
language: en
|
4 |
+
tags:
|
5 |
+
- sagemaker
|
6 |
+
- bert-base-uncased
|
7 |
+
- text classification
|
8 |
+
license: apache-2.0
|
9 |
+
datasets:
|
10 |
+
- adecorpusv2
|
11 |
+
model-index:
|
12 |
+
- name: BERT-ade_corpus
|
13 |
+
results:
|
14 |
+
- task:
|
15 |
+
name: Text Classification
|
16 |
+
type: text-classification
|
17 |
+
dataset:
|
18 |
+
name: "ade_corpus_v2Ade_corpus_v2_classification"
|
19 |
+
type: ade_corpus
|
20 |
+
|
21 |
+
---
|
22 |
+
|
23 |
+
## bert-base-uncased
|
24 |
+
This model was trained using Amazon SageMaker and the new Hugging Face Deep Learning container.
|
25 |
+
|
26 |
+
|
27 |
+
## Hyperparameters
|
28 |
+
```json
|
29 |
+
{
|
30 |
+
"do_eval": true,
|
31 |
+
"do_train": true,
|
32 |
+
"fp16": true,
|
33 |
+
"load_best_model_at_end": true,
|
34 |
+
"model_name": "bert-base-uncased",
|
35 |
+
"num_train_epochs": 10,
|
36 |
+
"per_device_eval_batch_size": 16,
|
37 |
+
"per_device_train_batch_size": 16,
|
38 |
+
"learning_rate":5e-5
|
39 |
+
|
40 |
+
}
|
41 |
+
```
|
42 |
+
## Validation Metrics
|
43 |
+
| key | value |
|
44 |
+
| --- | ----- |
|
45 |
+
| eval_accuracy | 0.9298021697511167 |
|
46 |
+
| eval_auc | 0.8902672664394546 |
|
47 |
+
| eval_f1 | 0.827315541601256 |
|
48 |
+
| eval_loss | 0.17835010588169098 |
|
49 |
+
| eval_recall | 0.8234375 |
|
50 |
+
| eval_precision | 0.831230283911672 |
|
51 |
+
|
52 |
+
"""
|