israelcamp
commited on
Commit
•
81864d3
1
Parent(s):
fe8c52b
Update README.md
Browse files
README.md
CHANGED
@@ -5,4 +5,47 @@ datasets:
|
|
5 |
- carolina-c4ai/corpus-carolina
|
6 |
language:
|
7 |
- pt
|
8 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
- carolina-c4ai/corpus-carolina
|
6 |
language:
|
7 |
- pt
|
8 |
+
---
|
9 |
+
|
10 |
+
|
11 |
+
# DeBERTinha XSmall (aka "debertinha-ptbr-xsmall")
|
12 |
+
|
13 |
+
## Introduction
|
14 |
+
|
15 |
+
DeBERTinha is a pretrained DeBERTa model for Brazilian Portuguese.
|
16 |
+
|
17 |
+
## Available models
|
18 |
+
|
19 |
+
| Model | Arch. | #Params |
|
20 |
+
| ---------------------------------------- | ---------- | ------- |
|
21 |
+
| `sagui-nlp/debertinha-ptbr-xsmall` | DeBERTa-V3-Xsmall | 40M |
|
22 |
+
|
23 |
+
## Usage
|
24 |
+
|
25 |
+
```python
|
26 |
+
from transformers import AutoTokenizer
|
27 |
+
from transformers import AutoModelForPreTraining
|
28 |
+
from transformers import AutoModel
|
29 |
+
|
30 |
+
model = AutoModelForPreTraining.from_pretrained('sagui-nlp/debertinha-ptbr-xsmall')
|
31 |
+
tokenizer = AutoTokenizer.from_pretrained('sagui-nlp/debertinha-ptbr-xsmall')
|
32 |
+
```
|
33 |
+
|
34 |
+
### For embeddings
|
35 |
+
|
36 |
+
```python
|
37 |
+
import torch
|
38 |
+
|
39 |
+
model = AutoModel.from_pretrained('sagui-nlp/debertinha-ptbr-xsmall')
|
40 |
+
input_ids = tokenizer.encode('Tinha uma pedra no meio do caminho.', return_tensors='pt')
|
41 |
+
|
42 |
+
with torch.no_grad():
|
43 |
+
outs = model(input_ids)
|
44 |
+
encoded = outs.last_hidden_state[0, 1:-1] # Ignore [CLS] and [SEP] special tokens
|
45 |
+
```
|
46 |
+
|
47 |
+
## Citation
|
48 |
+
|
49 |
+
If you use our work, please cite:
|
50 |
+
|
51 |
+
`Comming soon`
|