Update README.md

README.md

@@ -23,31 +23,63 @@ should probably proofread and complete it, then remove this comment. -->
 
 # dougtrajano/toxicity-target-classification
 
-This model is a fine-tuned version of [neuralmind/bert-base-portuguese-cased](https://huggingface.co/neuralmind/bert-base-portuguese-cased) on the OLID-BR dataset.
-It achieves the following results on the evaluation set:
-- Loss: 0.6110
-- Accuracy: 0.6864
-- F1: 0.6872
-- Precision: 0.6882
-- Recall: 0.6864
+Toxicity Target Classification is a model that classifies if a given text is targeted or not.
 
-## Model description
+This BERT model is a fine-tuned version of [neuralmind/bert-base-portuguese-cased](https://huggingface.co/neuralmind/bert-base-portuguese-cased) on the [OLID-BR dataset](https://huggingface.co/datasets/dougtrajano/olid-br).
 
-More information needed
+## Overview
 
-## Intended uses & limitations
+**Input:** Text in Brazilian Portuguese
 
-More information needed
+**Output:** Binary classification (targeted or untargeted)
 
-## Training and evaluation data
+## Usage
 
-More information needed
+```python
+from transformers import AutoTokenizer, AutoModelForSequenceClassification
+
+tokenizer = AutoTokenizer.from_pretrained("dougtrajano/toxicity-target-classification")
+
+model = AutoModelForSequenceClassification.from_pretrained("dougtrajano/toxicity-target-classification")
+```
+
+## Limitations and bias
+
+The following factors may degrade the model’s performance.
+
+**Text Language**:  The model was trained on Brazilian Portuguese texts, so it may not work well with Portuguese dialects.
+
+**Text Origin**: The model was trained on texts from social media and a few texts from other sources, so it may not work well on other types of texts.
+
+## Trade-offs
+
+Sometimes models exhibit performance issues under particular circumstances. In this section, we'll discuss situations in which you might discover that the model performs less than optimally, and should plan accordingly.
+
+**Text Length**: The model was fine-tuned on texts with a word count between 1 and 178 words (average of 18 words). It may give poor results on texts with a word count outside this range.
+
+## Performance
+
+The model was evaluated on the test set of the [OLID-BR](https://dougtrajano.github.io/olid-br/) dataset.
+
+**Accuracy:** 0.6864
+
+**Precision:** 0.6882
+
+**Recall:** 0.6864
+
+**F1-Score:** 0.6872
+
+| Class | Precision | Recall | F1-Score | Support |
+| :---: | :-------: | :----: | :------: | :-----: |
+| `UNTARGETED` | 0.4912 | 0.5011 | 0.4961 | 443 |
+| `TARGETED INSULT` | 0.7759 | 0.7688 | 0.7723 | 995 |
 
 ## Training procedure
 
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
+
 - learning_rate: 4.174021560583183e-05
 - train_batch_size: 8
 - eval_batch_size: 8
@@ -57,19 +89,13 @@ The following hyperparameters were used during training:
 - num_epochs: 30
 - label_smoothing_factor: 0.09936835309930625
 
-### Training results
-
-| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1     | Precision | Recall |
-|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
-| 0.633         | 1.0   | 537  | 0.6040          | 0.6919   | 0.5805 | 0.6351    | 0.6919 |
-| 0.5915        | 2.0   | 1074 | 0.6110          | 0.6864   | 0.6872 | 0.6882    | 0.6864 |
-| 0.4584        | 3.0   | 1611 | 0.7104          | 0.6933   | 0.6606 | 0.6605    | 0.6933 |
-| 0.3564        | 4.0   | 2148 | 0.9816          | 0.6168   | 0.6307 | 0.6671    | 0.6168 |
-
-
 ### Framework versions
 
 - Transformers 4.26.0
 - Pytorch 1.10.2+cu113
 - Datasets 2.9.0
 - Tokenizers 0.13.2
+
+## Provide Feedback
+
+If you have any feedback on this model, please [open an issue](https://github.com/DougTrajano/ToChiquinho/issues/new) on GitHub.