File size: 1,512 Bytes
e896a05
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
349a781
 
 
e896a05
 
 
 
 
 
 
 
 
 
 
 
 
349a781
 
 
e896a05
 
 
 
 
 
 
 
 
 
 
349a781
e896a05
349a781
e896a05
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
---
tags:
- autotrain
- text-classification
language:
- pt
widget:
- text: "I love AutoTrain 🤗"
datasets:
- alexandreteles/autotrain-data-told_br_binary_sm
co2_eq_emissions:
  emissions: 4.429755329718354
---

# Model Trained Using AutoTrain

- Problem type: Binary Classification
- Model ID: 2489276793
- Base model: BertForSequenceClassification
- Parameters: 109M
- Model size: 416MB
- CO2 Emissions (in grams): 4.4298

## Validation Metrics

- Loss: 0.432
- Accuracy: 0.800
- Precision: 0.823
- Recall: 0.704
- AUC: 0.891
- F1: 0.759

## Usage

This model was trained on a random subset of the [told_br](https://huggingface.co./datasets/told-br) dataset (1/3 of the original size). Our main objective is to provide a small
model that can be used to classify Brazilian Portuguese tweets in a binary way ('toxic' or 'non toxic').

You can use cURL to access this model:

```
$ curl -X POST -H "Authorization: Bearer YOUR_API_KEY" -H "Content-Type: application/json" -d '{"inputs": "I love AutoTrain"}' https://api-inference.huggingface.co/models/alexandreteles/autotrain-told_br_binary_sm-2489276793
```

Or Python API:

```
from transformers import AutoModelForSequenceClassification, AutoTokenizer

model = AutoModelForSequenceClassification.from_pretrained("alexandreteles/told_br_binary_sm", use_auth_token=True)

tokenizer = AutoTokenizer.from_pretrained("alexandreteles/told_br_binary_sm", use_auth_token=True)

inputs = tokenizer("I love AutoTrain", return_tensors="pt")

outputs = model(**inputs)
```