monsoon-nlp
commited on
Commit
·
2c4e641
1
Parent(s):
b91adf5
add readme and tags
Browse files
README.md
ADDED
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- en
|
4 |
+
- hi
|
5 |
+
- bn
|
6 |
+
- ta
|
7 |
+
- as
|
8 |
+
- gu
|
9 |
+
- kn
|
10 |
+
- ks
|
11 |
+
- ml
|
12 |
+
- mr
|
13 |
+
- ne
|
14 |
+
- or
|
15 |
+
- pa
|
16 |
+
- sa
|
17 |
+
- sd
|
18 |
+
- te
|
19 |
+
- ur
|
20 |
+
license: apache-2.0
|
21 |
+
---
|
22 |
+
|
23 |
+
## MuRIL - Unofficial
|
24 |
+
|
25 |
+
Multilingual Representations for Indian Languages : Google open sourced
|
26 |
+
this BERT model pre-trained on 17 Indian languages, and their transliterated
|
27 |
+
counterparts.
|
28 |
+
|
29 |
+
The model was trained using a self-supervised masked language modeling task. We do whole word masking with a maximum of 80 predictions. The model was trained for 1000K steps, with a batch size of 4096, and a max sequence length of 512.
|
30 |
+
|
31 |
+
Details: https://tfhub.dev/google/MuRIL/1
|
32 |
+
|
33 |
+
License: Apache 2.0
|
34 |
+
|
35 |
+
### About this upload
|
36 |
+
|
37 |
+
I ported the TFHub .pb model to .h5 and then pytorch_model.bin for
|
38 |
+
compatibility with Transformers.
|