https://github.com/savjee/yt-spam-classifier
Using TensorFlow to automatically filter crypto-related spam comments on YouTube.
https://github.com/savjee/yt-spam-classifier
Last synced: over 1 year ago
JSON representation
Using TensorFlow to automatically filter crypto-related spam comments on YouTube.
- Host: GitHub
- URL: https://github.com/savjee/yt-spam-classifier
- Owner: Savjee
- License: mit
- Created: 2021-06-16T14:11:35.000Z (about 5 years ago)
- Default Branch: master
- Last Pushed: 2024-12-20T20:16:01.000Z (over 1 year ago)
- Last Synced: 2025-03-18T01:51:27.358Z (over 1 year ago)
- Language: Jupyter Notebook
- Homepage:
- Size: 1.41 MB
- Stars: 38
- Watchers: 4
- Forks: 14
- Open Issues: 2
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# yt-spam-filter
Using TensorFlow to automatically filter crypto-related spam comments on YouTube.
🎬 See my video for more details: [https://youtu.be/zSEYC3CCA1I](https://youtu.be/zSEYC3CCA1I)
## Repository structure
* `00-CoreML test` -> First test with Apple's CoreML.
* `01-fetch-input-data` -> Python script that fetches comments from YouTube and stores them in a CSV file
* `02-rated-comments` -> Dataset of comments from my channel, marked as spam (1) or not spam (0)
* `03-tensorflow-training` -> Jupyter notebook used to train the text classifier with TensorFlow
* `04-inference` -> Python script that runs periodically to filter latest comments on my channel
## Installation
Clone this repo:
```
git clone https://github.com/Savjee/yt-spam-classifier.git
```
Install dependencies (might not be complete):
```
pip3 install -r requirements.txt
```
Create a `.env` file from the template, and fill in your YouTube API key and Channel ID
```
cp .env.dist .env
```
More instructions to follow.
## Contribute
Feel free to fork or submit improvements but be gentle. This is my first ML and Python project (it probably shows).