- Python >= 3.8
- obspy >=1.2.2
- pytorch >=1.8.1 (cu102)
- tensorboard >= 2.4.1 (for neural net demo)
- pytest >=6.2.2
- jupyter (or any other IPython notebook viewer)
- tqdm >=4.60.0 (used in demo/workflow notebooks)
Latest versions of all these dependencies should work fine. They may be installed using pip or whatever package management tool you might be using. If in doubt, use pip.
Once the dependencies are met, from the project root run pip install. I personally like editable installs but this is a personal preference.
eew-spatio-temporal$ pip install -e .
This will install est_lib.
All tests are written in Pytest. After installing dependencies, call pytest from the project root to run all tests. All tests should be passing and you should see an output similar to the snippet pasted below. In case of errors please confirm whether all dependencies are set up correct.
If problems still persist, please open an issue on this repository so that I can advise.
eew-spatio-temporal$ pytest
================================================= test session starts =================================================
platform win32 -- Python 3.8.8, pytest-6.2.4, py-1.10.0, pluggy-0.13.1
rootdir: C:\Users\aksha\Desktop\eew-spatio-temporal
collected 51 items
tests\test_dataset.py ........ [ 15%]
tests\test_event_parse.py ............... [ 45%]
tests\test_init.py . [ 47%]
tests\test_nn.py .......... [ 66%]
tests\test_nn_common.py .... [ 74%]
tests\test_util.py ............. [100%]
================================================= 51 passed in 22.27s =================================================
Refer to the IPython notebooks in the workflow_demos folder
for a tutorial on how est_lib might be used.
Refer to the folder est_lib/nn/demo to find two notebooks
that demo how the cp_line_simple neural network.