Commit
•
bc8e865
1
Parent(s):
6d1cdb6
Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
datasets:
|
3 |
+
- fake_news_english
|
4 |
+
language:
|
5 |
+
- en
|
6 |
+
library_name: transformers
|
7 |
+
---
|
8 |
+
|
9 |
+
# Model Card for Model ID
|
10 |
+
|
11 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
12 |
+
|
13 |
+
This modelcard aims to be a base template for new models. It has been generated using [this raw template](https://github.com/huggingface/huggingface_hub/blob/main/src/huggingface_hub/templates/modelcard_template.md?plain=1).
|
14 |
+
|
15 |
+
## Model Details
|
16 |
+
|
17 |
+
### Model Description
|
18 |
+
|
19 |
+
This model is created using the fake news dataset from kaggle. The custom model is a fine tuned distilbert model with additional layers.
|
20 |
+
The code was written in pytorch. The dataset was processed with removing symbols and converting text to lower case. The train - validate - test datasets
|
21 |
+
are created in the ratio 60:20:20. The model was trained for two epochs and obtained an accuracy of 99%.
|
22 |
+
However, the model has been shown to be overfitted on certain types of samples owing to lack of diversity in the samples. Please be cautious
|
23 |
+
before using this model for a downstream use case
|
24 |
+
|
25 |
+
- **Developed by:** Aishwarya A. Nair
|
26 |
+
- **Shared by [optional]:** [More Information Needed]
|
27 |
+
- **Model type:** [More Information Needed]
|
28 |
+
- **Language(s) (NLP):** [More Information Needed]
|
29 |
+
- **License:** [More Information Needed]
|
30 |
+
- **Finetuned from model [optional]:** distilbert-base-uncased
|
31 |
+
|
32 |
+
### Model Sources [optional]
|
33 |
+
|
34 |
+
<!-- Provide the basic links for the model. -->
|
35 |
+
|
36 |
+
- **Notebook:** [https://colab.research.google.com/drive/1G65Ye1UC-QeQXAJN9WPkGHvM0Qgo4Mf5?usp=sharing]
|
37 |
+
|
38 |
+
## Uses
|
39 |
+
Fake news detection can be used in the cases when you need to verify the veracity of a news article or a tweet or other pieces of text.
|
40 |
+
|
41 |
+
## Bias, Risks, and Limitations
|
42 |
+
|
43 |
+
The model has been shown to be overfitted on certain types of samples owing to lack of diversity in the samples. Please be cautious
|
44 |
+
before using this model for a downstream use case
|
45 |
+
|