Update README.md
Browse files
README.md
CHANGED
@@ -46,67 +46,6 @@ if __name__ == "__main__":
|
|
46 |
predict()
|
47 |
```
|
48 |
|
49 |
-
## β
Install requirements
|
50 |
-
|
51 |
-
Install required dependencies
|
52 |
-
|
53 |
-
```sh
|
54 |
-
pip install --upgrade pip
|
55 |
-
pip install -r requirements.txt
|
56 |
-
```
|
57 |
-
|
58 |
-
## β
Add BERT virtual env
|
59 |
-
|
60 |
-
write the command below
|
61 |
-
|
62 |
-
```sh
|
63 |
-
# β
Create and activate a virtual environment
|
64 |
-
python -m venv bert-env
|
65 |
-
source bert-env/bin/activate # On Windows use: bert-env\Scripts\activate
|
66 |
-
```
|
67 |
-
|
68 |
-
## β
Install CUDA
|
69 |
-
|
70 |
-
Check if your GPU supports CUDA:
|
71 |
-
|
72 |
-
```sh
|
73 |
-
nvidia-smi
|
74 |
-
```
|
75 |
-
|
76 |
-
Then:
|
77 |
-
|
78 |
-
```sh
|
79 |
-
pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu121
|
80 |
-
PYTORCH_CUDA_ALLOC_CONF=expandable_segments:False
|
81 |
-
```
|
82 |
-
|
83 |
-
## π§ How to use
|
84 |
-
|
85 |
-
- Check your device and CUDA availability:
|
86 |
-
|
87 |
-
```sh
|
88 |
-
python check_device.py
|
89 |
-
```
|
90 |
-
|
91 |
-
> :warning: Using CPU is not advisable, prefer check your CUDA availability.
|
92 |
-
|
93 |
-
- Train the model:
|
94 |
-
|
95 |
-
```sh
|
96 |
-
python scripts/train.py
|
97 |
-
```
|
98 |
-
|
99 |
-
> :warning: Remove unneeded checkpoint in models/pretrained to save your storage after training
|
100 |
-
|
101 |
-
- Run prediction:
|
102 |
-
|
103 |
-
```sh
|
104 |
-
python scripts/predict.py
|
105 |
-
```
|
106 |
-
|
107 |
-
β
Dataset Location: [`data/spam.csv`](./data/spam.csv), modify the dataset to enhance the model based on your needs.
|
108 |
-
|
109 |
-
|
110 |
## π Citations
|
111 |
|
112 |
If you use this repository or its ideas, please cite the following:
|
@@ -124,13 +63,3 @@ See [`citations.bib`](./citations.bib) for full BibTeX entries.
|
|
124 |
- Logging silencing inspired by Hugging Face GitHub discussions
|
125 |
- Dataset from [UCI SMS Spam Collection](https://www.kaggle.com/datasets/uciml/sms-spam-collection-dataset)
|
126 |
- Inspiration from [Kaggle Notebook by Suyash Khare](https://www.kaggle.com/code/suyashkhare/naive-bayes)
|
127 |
-
|
128 |
-
## License and Usage
|
129 |
-
|
130 |
-
License under [MIT license](./LICENSE).
|
131 |
-
|
132 |
-
---
|
133 |
-
|
134 |
-
Leave a β if you think this project is helpful, contributions are welcome.
|
135 |
-
|
136 |
-
---
|
|
|
46 |
predict()
|
47 |
```
|
48 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
49 |
## π Citations
|
50 |
|
51 |
If you use this repository or its ideas, please cite the following:
|
|
|
63 |
- Logging silencing inspired by Hugging Face GitHub discussions
|
64 |
- Dataset from [UCI SMS Spam Collection](https://www.kaggle.com/datasets/uciml/sms-spam-collection-dataset)
|
65 |
- Inspiration from [Kaggle Notebook by Suyash Khare](https://www.kaggle.com/code/suyashkhare/naive-bayes)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|