Commit
•
1e2c2b1
1
Parent(s):
1c6be11
Upload README.md
Browse files
README.md
CHANGED
@@ -8,27 +8,15 @@ library_name: transformers
|
|
8 |
pipeline_tag: text-classification
|
9 |
---
|
10 |
|
11 |
-
##
|
12 |
-
|
13 |
-
|
14 |
-
## How to use
|
15 |
-
|
16 |
-
You can use the model with the following code.
|
17 |
|
18 |
-
|
19 |
-
|
20 |
|
21 |
-
|
22 |
-
|
23 |
-
model = BertForSequenceClassification.from_pretrained(model_path, num_labels=2)
|
24 |
|
25 |
-
pipeline = TextClassificationPipeline(model=model, tokenizer=tokenizer)
|
26 |
-
print(pipeline("You're a fucking nerd."))
|
27 |
-
```
|
28 |
|
29 |
## Training data
|
30 |
The training data comes from this [Kaggle competition](https://www.kaggle.com/c/jigsaw-unintended-bias-in-toxicity-classification/data). We use 90% of the `train.csv` data to train the model.
|
31 |
-
|
32 |
-
## Evaluation results
|
33 |
-
|
34 |
-
The model achieves 0.95 AUC in a 1500 rows held-out test set.
|
|
|
8 |
pipeline_tag: text-classification
|
9 |
---
|
10 |
|
11 |
+
## Name
|
12 |
+
Vishwa Patel
|
|
|
|
|
|
|
|
|
13 |
|
14 |
+
## Project
|
15 |
+
Toxic Comment Classification
|
16 |
|
17 |
+
## Model description
|
18 |
+
This model is a fine-tuned version of the [bert-base-uncased](https://huggingface.co/transformers/model_doc/bert.html) model to classify toxic comments.
|
|
|
19 |
|
|
|
|
|
|
|
20 |
|
21 |
## Training data
|
22 |
The training data comes from this [Kaggle competition](https://www.kaggle.com/c/jigsaw-unintended-bias-in-toxicity-classification/data). We use 90% of the `train.csv` data to train the model.
|
|
|
|
|
|
|
|