https://github.com/nedap/dev-day-rfidtags
Nedap DevDay 2019 retail case: automatic RFID label classification using machine learning
https://github.com/nedap/dev-day-rfidtags
Last synced: 3 months ago
JSON representation
Nedap DevDay 2019 retail case: automatic RFID label classification using machine learning
- Host: GitHub
- URL: https://github.com/nedap/dev-day-rfidtags
- Owner: nedap
- Created: 2019-12-02T14:31:52.000Z (over 6 years ago)
- Default Branch: master
- Last Pushed: 2022-11-28T11:53:53.000Z (over 3 years ago)
- Last Synced: 2025-03-25T04:07:54.570Z (about 1 year ago)
- Language: Jupyter Notebook
- Size: 3.9 MB
- Stars: 1
- Watchers: 3
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# DevDay 2019 - Retail RFID Label Scanning
In this data science mini-hackathon we are going to work with a real-life problem from Nedap Retail. In the world of RFID label scanning, it is very hard to isolate products since the entire store is filled with RFID tags (see Figure 1 and Figure 2). Our challenge is to come up with an algorithm to do this separation automatically when scanning a batch of newly arriving products.
You are going to develop and evaluate a basic machine learning pipeline to address the case above. We give you all the tools needed to get started with basic machine learning to show you the power of these algorithms.
**Figure 1: scanning RFID labels in a typical retail store**
**Figure 2: label scanning can be noisy**
## Environment
First, you can clone this repository to a location of your choice:
```sh
git clone https://github.com/nedap/dev-day-rfidtags.git
```
If you don't have git installed, you can also download the repository [here](https://github.com/nedap/dev-day-rfidtags/archive/master.zip).
Afterwards, we will setup the environment.
### Option 1: Anaconda
Download anaconda here: [installation instructions](https://www.anaconda.com/distribution/#download-section).
#### Linux/macOS
Afterwards, run the following commands in the root of this repository.
```sh
conda env update -f environment.yml
conda activate rfidtags
pip install --quiet ipykernel autopep8
python -m ipykernel install --user --name rfidtags --display-name "Python (rfidtags)"
jupyter notebook
```
Jupyter should automatically open a page in your browser.
#### Windows
Open the Anaconda Navigator. Then, create a new virtual environment called `rfidtags` by importing the `environment.yml` in this repository.
Once this is done, select that environment and launch Jupyter.
### Option 2: Docker
Make sure you have docker installed: [installation instructions](https://docs.docker.com/v17.09/engine/installation/#supported-platforms).
Start the docker container and visit http://localhost:8888/notebooks/work/notebooks/rfid-classification.ipynb (token: `devday`).
```sh
cd dev-day-rfidtags/
docker-compose up
```