migueladarlo
commited on
Commit
•
8afaaaf
1
Parent(s):
22ba177
Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,57 @@
|
|
1 |
---
|
2 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
+
language:
|
3 |
+
- en
|
4 |
+
license: mit # Example: apache-2.0 or any license from https://huggingface.co/docs/hub/model-repos#list-of-license-identifiers
|
5 |
+
tags:
|
6 |
+
- text # Example: audio
|
7 |
+
- Twitter
|
8 |
+
datasets:
|
9 |
+
- CLPsych 2015 # Example: common_voice. Use dataset id from https://hf.co/datasets
|
10 |
+
metrics:
|
11 |
+
- accuracy, f1, precision, recall, AUC # Example: wer. Use metric id from https://hf.co/metrics
|
12 |
+
|
13 |
+
model-index:
|
14 |
+
- name: distilbert-depression-base
|
15 |
+
results: []
|
16 |
---
|
17 |
+
|
18 |
+
# distilbert-depression-base
|
19 |
+
|
20 |
+
This model is a fine-tuned version of [base-uncased](https://huggingface.co/distilbert-base-uncased) trained on CLPsych 2015 and evaluated on a scraped dataset from Twitter.
|
21 |
+
It achieves the following results on the evaluation set:
|
22 |
+
- Evaluation Loss: 0.64
|
23 |
+
- Accuracy: 0.65
|
24 |
+
- F1: 0.70
|
25 |
+
- Precision: 0.61
|
26 |
+
- Recall: 0.83
|
27 |
+
- AUC: 0.65
|
28 |
+
|
29 |
+
|
30 |
+
## Intended uses & limitations
|
31 |
+
|
32 |
+
Feed a corpus of tweets to the model to generate label if input is indicative of depression or not.
|
33 |
+
|
34 |
+
Limitation: All token sequences longer than 512 are automatically truncated.
|
35 |
+
|
36 |
+
## Training hyperparameters
|
37 |
+
|
38 |
+
The following hyperparameters were used during training:
|
39 |
+
- learning_rate: 3.39e-05
|
40 |
+
- train_batch_size: 16
|
41 |
+
- eval_batch_size: 16
|
42 |
+
- weight_decay: 0.13
|
43 |
+
- num_epochs: 3.0
|
44 |
+
|
45 |
+
## Training results
|
46 |
+
|
47 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
48 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
49 |
+
| 0.68 | 1.0 | 625 | 0.1385 | 0.9745 |
|
50 |
+
| 0.60 | 2.0 | 1250 | 0.1385 | 0.9745 |
|
51 |
+
| 0.52 | 3.0 | 1875 | 0.1385 | 0.9745 |
|
52 |
+
|
53 |
+
| Epoch | Training Loss | Validation Loss | Accuracy | F1 | Precision | Recall | AUC |
|
54 |
+
|:-----:|:-------------:|:---------------:|:--------:|:--------:|:---------:|:--------:|:--------:|
|
55 |
+
| 1.0 | 0.68 | 0.66 | 0.59 | 0.63 | 0.56 | 0.73 | 0.59 |
|
56 |
+
| 2.0 | 0.60 | 0.68 | 0.63 | 0.69 | 0.59 | 0.83 | 0.63 |
|
57 |
+
| 3.0 | 0.52 | 0.67 | 0.64 | 0.66 | 0.62 | 0.72 | 0.65 |
|