sl
commited on
Commit
•
9c0740f
1
Parent(s):
d78e775
init
Browse files- README.md +30 -1
- config.json +30 -0
- pytorch_model.bin +3 -0
- vocab.txt +0 -0
README.md
CHANGED
@@ -1,3 +1,32 @@
|
|
1 |
---
|
2 |
-
|
|
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
+
language:
|
3 |
+
- zh
|
4 |
+
license: "apache-2.0"
|
5 |
---
|
6 |
+
|
7 |
+
## Chinese-MobileBERT
|
8 |
+
> The original [Chinese-MobileBERT](https://github.com/ymcui/Chinese-MobileBERT) repository does not provide pytorch weights, here the weights are converted via the [model_convert](https://github.com/CycloneBoy/model_convert) repository.
|
9 |
+
|
10 |
+
This repository is developed based on:https://github.com/ymcui/Chinese-MobileBERT
|
11 |
+
|
12 |
+
You may also be interested in,
|
13 |
+
- Chinese BERT series: https://github.com/ymcui/Chinese-BERT-wwm
|
14 |
+
- Chinese MacBERT: https://github.com/ymcui/MacBERT
|
15 |
+
- Chinese ELECTRA: https://github.com/ymcui/Chinese-ELECTRA
|
16 |
+
- Chinese XLNet: https://github.com/ymcui/Chinese-XLNet
|
17 |
+
- Knowledge Distillation Toolkit - TextBrewer: https://github.com/airaria/TextBrewer
|
18 |
+
|
19 |
+
More resources by HFL: https://github.com/ymcui/HFL-Anthology
|
20 |
+
|
21 |
+
## Citation
|
22 |
+
If you find the technical report or resource is useful, please cite the following technical report in your paper.
|
23 |
+
|
24 |
+
|
25 |
+
```
|
26 |
+
@misc{cui-2022-chinese-mobilebert,
|
27 |
+
title={Chinese MobileBERT},
|
28 |
+
author={Cui, Yiming},
|
29 |
+
howpublished={\url{https://github.com/ymcui/Chinese-MobileBERT}},
|
30 |
+
year={2022}
|
31 |
+
}
|
32 |
+
```
|
config.json
ADDED
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "cycloneboy/chinese_mobilebert_base_f2",
|
3 |
+
"architectures": [
|
4 |
+
"MobileBertForPreTraining"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"hidden_act": "relu",
|
8 |
+
"hidden_dropout_prob": 0.0,
|
9 |
+
"hidden_size": 768,
|
10 |
+
"initializer_range": 0.02,
|
11 |
+
"intermediate_size": 512,
|
12 |
+
"key_query_shared_bottleneck": true,
|
13 |
+
"layer_norm_eps": 1e-12,
|
14 |
+
"max_position_embeddings": 512,
|
15 |
+
"num_attention_heads": 12,
|
16 |
+
"num_hidden_layers": 12,
|
17 |
+
"embedding_size": 768,
|
18 |
+
"trigram_input": true,
|
19 |
+
"use_bottleneck": true,
|
20 |
+
"intra_bottleneck_size": 144,
|
21 |
+
"use_bottleneck_attention": true,
|
22 |
+
"num_feedforward_networks": 2,
|
23 |
+
"normalization_type": "no_norm",
|
24 |
+
"pad_token_id": 0,
|
25 |
+
"model_type": "mobilebert",
|
26 |
+
"transformers_version": "4.6.0.dev0",
|
27 |
+
"classifier_activation": false,
|
28 |
+
"type_vocab_size": 2,
|
29 |
+
"vocab_size": 21128
|
30 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f0d4a2b2143d2d60b9b00758bfe1be04234d6065b2ace034a0bc680810282a3d
|
3 |
+
size 110552707
|
vocab.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|