image-embedding-blog

Image Embeddings Blog

Supplementary code for our blog on Pre-Trained Image Embeddings with HuggingFace.

Pre-requisites

You'll need following pre-requisites to run the demo:

• Python 3
• Conda/Miniconda

Getting started

1. Clone this repo:

git clone git@github.com:fuzzylabs/tech-deep-dive-blogs.git

2. Go to the cloned directory:

cd image-embeddings-blog

3. Create a new Python virtual environment with Conda, activate it and install required libraries. For Linux/MacOS users:

conda create -y --name demoenv python==3.9.0
conda activate demoenv
pip install -r requirements.txt

When you are finished with this tutorial you exit the Conda environment using:

conda deactivate

The data directory contains all images required for this tutorial.

Use the embbeddings.ipynb notebook to follow along with the blog.

ZenML Pipeline

The code within zenml-pipeline can be run using the following steps:

Initialise ZenML:

$ zenml init

Create ZenML server:

$ zenml up

Run the pipeline from the image-embedding-blog directory:

$ python3 zenml-pipeline/run.py