CS498 Spring 2021: AI for Robotic Manipulation
Instructor: Kris Hauser
For the first few assignments, the most convenient way to get started is through Jupyter Notebook is through this Binder environment. Any time the class repository is changed, the first unlucky person to log on to Binder will encounter a lengthy build process that will take several minutes to complete. Subsequent logins will be much faster (about 20 seconds).
Once the environment launches, just navigate to the MPX/MPX.ipynb notebook in the appropriate folder to get started.
If you are working in Binder, make sure to save your work to your local machine before closing the browser window. To do so, click on the "Save to browser storage" / "Restore from browser storage" buttons.
To download to your local machine for submission, click the "Download" button. To upload a notebook or code files, go back to the Jupyter file browser and choose the "Upload" button.
Local installation should be possible on Linux, Windows, and Mac. Instructions are as follows:
Install Python 3.x
Install Python packages. On Linux / Mac this can be as easy as using pip
python -m pip install klampt PyOpenGL numpy scipy scikit-learn matplotlib notebook
For Windows, you may need to download the install files from this repository.
Install the Klampt-jupyter-extension package.
git clone https://github.com/krishauser/Klampt-jupyter-extension.git cd Klampt-jupyter-extension/jupyter-nbextension jupyter nbextension install klampt/ --user jupyter nbextension enable klampt/main jupyter nbextension enable klampt/three.min jupyter nbextension enable klampt/KlamptFrontend cd ../..
You will be asked to periodically retrieve updated assignments; this will be done using
To launch Jupyter, open a command line terminal, navigate to your
cs498ir_s2021 directory, and enter:
Once the environment launches, you may navigate to the MPX/MPX.ipynb notebook in the appropriate folder to get started.