-
Notifications
You must be signed in to change notification settings - Fork 1
[WIP] Add Docker environment and testing the scripts #4
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: main
Are you sure you want to change the base?
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Nice! 🐳
@@ -92,35 +94,35 @@ | |||
np.savez_compressed(wd / "cubesxy" / f"{geojson.stem}.npz", X=cdata.astype("uint16"), attrs=data.imagery.attrs) | |||
continue |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@srmsoumya Looks like the code after the continue
wont run anymore.
environment.yml
Outdated
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@Rub21 I think we should clean up the environment.yml
file to packages we are using in the repository.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@srmsoumya yes, it would be good, I set that because i did not want to miss any of the libraries that you used for training the module.
I rebased this PR from main after changing some file names. Hope I did not break anything. @Rub21 and @srmsoumya you might have to reset your local branches after this (hope I did not create too much confusion with that. Otherwise feel free to undo my rebase) |
@srmsoumya @yellowcap, I am in the process of recreating the LULC module training, but I'm encountering issues with recreating the same data generations that you've successfully uploaded to s3. Firstly, I've created a Docker file that installs all the modules Soumya used for module training. The file can be found here: https://github.com/developmentseed/time-series-for-lulc/blob/env/docker/environment.yml Secondly, I've executed the data generation script sequentially, following Soumya's guidance. To verify the consistency of my results with Soumya's, I implemented an MD5 comparison script. However, my results aren't aligning with Soumya's data from S3. In fact, many of the files do not even exist. This discrepancy has led me to question whether you've used all the training geojson files from I'd appreciate any insights you could provide to help resolve these issues.. In case it is required , I am uploading the diles to this folder: s3://ds-labs-lulc/rub21/stacks/
s3://ds-labs-lulc/rub21/cubesxy/ Also the Instruction are here: https://github.com/developmentseed/time-series-for-lulc/blob/env/docker/start.sh |
Adding docker environment in order to run scripts and notebooks using the right version of the dependencies.
cc. @srmsoumya @batpad @yellowcap