https://www.kaggle.com/savasy/ttc4900

Arslan fe3e40ac67 Merge from Dev: Update Docker config, use gunicorn and remove Pytorch 1 month ago
.dvc 3835ed1044 Add base data and track it with DVC. Add associated python requirements 1 month ago
artifacts 8983e96d17 Feature: Logistic Classifier 1 month ago
data 8983e96d17 Feature: Logistic Classifier 1 month ago
metrics a8aba98742 Add necessary folders - data,metrics,artifacts 1 month ago
news_cat fe3e40ac67 Merge from Dev: Update Docker config, use gunicorn and remove Pytorch 1 month ago
scripts 8983e96d17 Feature: Logistic Classifier 1 month ago
.dockerignore fe3e40ac67 Merge from Dev: Update Docker config, use gunicorn and remove Pytorch 1 month ago
.dvcignore 3835ed1044 Add base data and track it with DVC. Add associated python requirements 1 month ago
.gitignore 92fdede2b3 Add basic app config (shared) 1 month ago
Dockerfile fe3e40ac67 Merge from Dev: Update Docker config, use gunicorn and remove Pytorch 1 month ago
LICENSE 6ee8331777 Initial commit 1 month ago
README.md 4bd0544299 Add Dockerfile, and related build instructions in README 1 month ago
app.py f352145576 Add api base/skeleton 1 month ago
docker-build.sh 4bd0544299 Add Dockerfile, and related build instructions in README 1 month ago
params.yaml 8983e96d17 Feature: Logistic Classifier 1 month ago
poetry.lock fe3e40ac67 Merge from Dev: Update Docker config, use gunicorn and remove Pytorch 1 month ago
pyproject.toml fe3e40ac67 Merge from Dev: Update Docker config, use gunicorn and remove Pytorch 1 month ago
requirements.txt fe3e40ac67 Merge from Dev: Update Docker config, use gunicorn and remove Pytorch 1 month ago

Data Pipeline

Legend
DVC Managed File
Git Managed File
Metric
Stage File
External File

README.md

Turkish Text Categorization

This is meant to serve as an example of how to structure and work on your ML projects, and deploy the resultant models.

The dataset I have chosen can be found on Kaggle.

  • The python packages can be installed via the requirements.txt file (in a venv), or using Poetry (preferred way).
  • To get the models and data files, you'll also need DVC. Just run dvc pull in this repo to get the data/artifacts.

Docker

  • The docker images can be built easily using the docker-build.sh file (you can change the tag name if you want).
  • Then run the image simply using docker run -d -p (your host machine port):8080 newscat (or the other name)
  • The port on docker can be configured using the env variable port.

Examples:

docker run -d -p 8085:8080 newscat

docker run -it -p 8085:8081 --env PORT=8081 newscat