Skip to content
Sample code used for explaining Pymapd API usage.
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
README.md
flights_depdelay.ipynb
flights_depdelay_cudf.ipynb
pymapd_kinesis_consumer.ipynb
pymapd_usage.ipynb

README.md

pymapd-workshop

Sample code used for explaining Pymapd API usage.

Here are the steps to create an Anaconda Python environment, and install Pymapd with the basic Machine Learning packages. This setup information applies to Linux Centos 7.

sudo yum install bzip2
curl -O https://repo.anaconda.com/archive/Anaconda3-5.3.1-Linux-x86_64.sh
bash Anaconda3-5.3.1-Linux-x86_64.sh
Confirm changes to ~/.bashrc and source it.
source .bashrc
Install Pymapd, Nvidia CUDA/RAPIDS along with the ML packages like scikit-learn and xgboost.
conda install -c nvidia/label/cuda10.0 -c rapidsai/label/cuda10.0 -c numba -c conda-forge -c defaults cudf=0.6 pymapd python=3.6
conda install scikit-learn
conda install -c conda-forge xgboost
conda install cudatoolkit
conda install -c anaconda statsmodels
Generate the Jupyter Notebook configuration file under ~/.jupyter/jupyter_notebook_config.py.
The config parameters c.NotebookApp.port and c.NotebookApp.ip control the port and IP from where you can connect. The default is port 8888.
jupyter notebook --generate-config 
You can’t perform that action at this time.