https://github.com/marcus-24/stock-predictor-ml-training
Machine Learning Operations (MLOps) pipeline for my stock prediction forecast model
https://github.com/marcus-24/stock-predictor-ml-training
hugging-face neptune-ai tensorflow time-series-forecasting yahoo-finance
Last synced: 3 months ago
JSON representation
Machine Learning Operations (MLOps) pipeline for my stock prediction forecast model
- Host: GitHub
- URL: https://github.com/marcus-24/stock-predictor-ml-training
- Owner: marcus-24
- Created: 2025-01-03T23:27:00.000Z (9 months ago)
- Default Branch: main
- Last Pushed: 2025-06-28T19:43:22.000Z (4 months ago)
- Last Synced: 2025-06-28T20:34:46.808Z (4 months ago)
- Topics: hugging-face, neptune-ai, tensorflow, time-series-forecasting, yahoo-finance
- Language: Python
- Homepage:
- Size: 76.2 KB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Stock-Predictor-ML-Training
## Objective
This repository is used to train the Tensorflow model, track the training metrics on Neptune AI, and deploy model artifacts to Hugging Face if the model accuracy has improved since the deployed model became stale.
## Project Setup
### Python Environment
First install the Python environment using:
`conda env create -f environment.yml`
Then you can activate the environment with the command below:
`conda activate stock_ml_train`
### Neptune AI Experiment Tracker
To track each training iteration, you will first need to create an account at Neptune.ai. Then follow the Create a Neptune Project tutorial to create project that will store each experiment run.
In order for this code to communicate with Neptune.ai, you will need to provide it the API token as shown in the Set Neptune credentials page. This code requires the user to store the token as `NEPTUNE_TOKEN` in a `.env` file.
### Hugging Face Credentials
## Connected Services
This repository interacts with the following services below: