search

jakewalter / easyQuake

Simplified machine-learning driven earthquake detection, location, and analysis.
MIT License
63 stars 21 forks source link

Issue Running Examples #3

Open Bryanrt-geophys opened 3 years ago

Bryanrt-geophys commented 3 years ago

Hello Jake,

I am attempting to run easyQuake and I have ran into a few preliminary challenges. One is though I have conda forge as an accessible channel, I cannot seem to download tensorflow-gpu==2.1.

Secondly, I do not have a machine with a GPU. Your write up does not explicitly state that a gpu is necessary. Can easyQuake be ran with a CPU?

Third, I am attempting to work through the idaho_example.py and I receive an error for the project_folder variable in the download_mseed(), detection_continous(), and association_sontinous() functions. When I try to run these this is what is returned: OSError: [Errno 30] Read-only file system: '/data'. Tho pointers "data/id" where pre-populated in the file, is this something I should change to a directory that exists or is this used to create a new directory to populate the downloaded files?

Bryan

jakewalter commented 3 years ago

  1. You need to make sure that the conda environment is python 3.7, higher python versions will not have tensorflow-gpu==2.1 available. Future work is intended to bring everything up to compatibility with python 3.9 and the latest tensorflow.

  2. The example codes assume that you are running a system with NVIDIA CUDA drivers and tensorflow-gpu installed. CPU is possible but I admittedly have not spent much time on thoroughly testing.

  3. Define the full path on your system - that is the one that is tested I am not sure whether it handles relative paths very well.