Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/flysirin/adstextclassification
Classification of advertisements by topic
https://github.com/flysirin/adstextclassification
docker excel flask pandas python pytorch scikit-learn
Last synced: about 1 month ago
JSON representation
Classification of advertisements by topic
- Host: GitHub
- URL: https://github.com/flysirin/adstextclassification
- Owner: flysirin
- Created: 2024-09-03T09:12:20.000Z (4 months ago)
- Default Branch: main
- Last Pushed: 2024-09-11T14:37:04.000Z (4 months ago)
- Last Synced: 2024-10-14T03:40:51.838Z (3 months ago)
- Topics: docker, excel, flask, pandas, python, pytorch, scikit-learn
- Language: Python
- Homepage:
- Size: 11.3 MB
- Stars: 1
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Text Classification with PyTorch and Scikit-Learn
This project focuses on text classification using a neural network built with PyTorch. It leverages `CountVectorizer` from `scikit-learn` for text preprocessing and includes evaluation metrics such as accuracy, precision, recall, and F1-score to measure the model’s performance.
Contains [AppFlaskPyTorch](AppFlaskPyTorch) - server with Flask and PyTorch on docker, click for read more information.
## How to use
1. Put labeled data **excel** file in `./data/train_data` for train model
Column with text - named - 'text'
Column with label - named - 'label'
2. Run train model
```shell
python train.py
```
3. Model and parameters save in `./saved/model/`
4. Put your data witch you want to classificate in `./data/for_prediction_data`
5. ```python run_predict_data.py```
6. Classificated data will be saved in `./data/output_data` with `label`
## Requirements
```shell
pip install -r requirements.txt
```
##### Contains
```
torch==2.4.0
pandas==2.2.2
scikit-learn==1.5.1
openpyxl
torchtext==0.18.0
numpy<2
```
`numpy<2 ` - downgrade version for exclude conflicts
## Special requirements for using Cuda
```
pip install -r requirements_cuda.txt
```
##### Contains
```
# Specifying the source for PyTorch packages
--index-url https://download.pytorch.org/whl/cu121
# Packages with specific versions and source
torch==2.4.1
torchvision==0.19.1
torchaudio==2.4.1
# Packages from standard PyPI
--extra-index-url https://pypi.org/simple
pandas==2.1.2
scikit-learn==1.5.1
openpyxl
torchtext==0.18.0
```
#### Further development in progress