File size: 468 Bytes
33265fb
 
56f4e90
 
8928304
33265fb
56f4e90
 
2e735aa
56f4e90
2e735aa
56f4e90
583c00d
56f4e90
f1b36b5
 
 
 
 
 
2e735aa
f1b36b5
2e735aa
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
---
license: creativeml-openrail-m

datasets:
- FredZhang7/toxi-text-3M
---


Estimated Time of Arrival (ETA): 6/30/2023

Estimated Final Train & Validation Accuracy: 96-98%

Large model (v2) will be avaliable for PyTorch

Lightweight model and tokenizer (v1) will be avaliable for transformers.js

<br>

<br>

Models tested: roberta, xlm-roberta, bert-tiny, bert-base-cased/uncased, bert-multilingual-cased/uncased, alberta-large-v2

Decision: bert-multilingual-cased