Skip to content


Repository files navigation

Yews Logo

Yews is a deep learning toolbox for processing seismic waveform made with flexibility, speed, and usability in mind. It is built upon PyTorch for researchers interested in applying deep learning techniques on seismic waveform data.


To ensure the GPU-powered PyTorch , first isntall PyTorch using the offical guide: and then install Yews via one of the following approaches:


conda install -c lijunzhu -c pytorch yews


conda install -c conda-forge yews


pip install yews

From source:

python install


  1. Running the above command without first installing PyTorch may still work. Depending on the OS, you may get either the GPU or CPU version of PyTorch. For example, MacOS currently will get the CPU Pytorch while Linux will get the GPU PyTorch by default. Please refer to for details.

  2. yews by itself is noarch, which means it is pure Python and OS independent. Most inconsistenciews between OS's are primarily due to the upstream difference (e.g. PyTorch and NumPy).

  3. obspy is an optional dependency, which is used for seismic waveform I/O; however, yews's core functionalities do not depend on obspy.

  4. You can install all yews optional dependencies via pip install yews[all].

  5. Below are the instructions to each optional dependencies to install them separately in conda.




    conda install -c conda-forge obspy


    conda install scipy


    conda install tqdm


You can find the API documentation on the yews website:


We appreciate all contributions. If you are planning to contribute back bug-fixes, please do so without any further discussion. If you plan to contribute new features, utility functions or extensions, please first open an issue and discuss the feature with us.