Added sample for geospatial classification #7291
Conversation
There was a problem hiding this comment.
Hi Nikita! Thanks for taking the time to collaborate with this sample, we really appreciate it!
This is looking great! I made some comments, mostly on style and nitpicking.
Some background: I learned C and C++ in college and I grew writing imperative code, but the more that I've been delving into purely functional languages (like Elm and Haskell), I've been learning really cool ways of doing the same things but without side effects like mutating variables. I've noticed it's usually shorter, more concise, easier to read, and less error prone, but it does require a slightly different way of doing things. I think in summary, avoiding to mutate variables, avoiding for-loops and while-loops, and using list comprehensions or functions instead. I've reinforced this with Beam / Dataflow, Earth Engine, and even TensorFlow Datasets, where everything is done via
map,filter, and some kind ofreduce, and it makes it easier to parallelize / distribute.
people-and-planet-ai/geospatial-classification/requirements.txt
Outdated
Show resolved
Hide resolved
people-and-planet-ai/geospatial-classification/requirements-test.txt
Outdated
Show resolved
Hide resolved
There was a problem hiding this comment.
This is looking great! I left some minor comments on the notebook. In general, can we keep the cell outputs of the code cells? That way people can skim through the notebook and see the outputs without having to actually run it. Although we should remove any cell output that includes personal information like a project ID or a bucket name.
Thanks David! Just resolved your last few comments :) |
|
All tests are passing, I think this should be ready to merge. Thanks @nikitamaia for contributing in this lovely sample! |
| "ignored_versions": ["2.7", "3.6", "3.7", "3.9", "3.10"], | ||
| # Old samples are opted out of enforcing Python type hints | ||
| # All new samples should feature them | ||
| # "enforce_type_hints": True, |
There was a problem hiding this comment.
We do ask that new samples use type hints as per the authoring guide https://github.com/GoogleCloudPlatform/python-docs-samples/blob/main/AUTHORING_GUIDE.md#documenting-types - please switch that to true and add them in
Description
Geospatial classification sample for the People + Planet AI series.