GEEpython download image unable to detect task progress on GEE code editor

[zwy1502]

In the past few weeks, have any classmates been unable to monitor task progress in the GEE code editor while downloading GEE resources using GEEpython? I have been encountering this problem recently, which is that whether I use GEemap or GEEpython's task. start() method to download GEE resources, there will always be a situation where the task cannot be seen in the GEE code editor. However, after a period of time, the resource will indeed appear on my Google Cloud Drive as scheduled. Very strange. It will eventually be downloaded to my Google Cloud drive, but the task progress cannot be seen in the GEE code editor.

[giswqs]

If this is the case, it is probably a Python API issue rather than a geemap issue

[giswqs]

I can't reproduce the issue. It works fine on my end.

[zwy1502]

That's strange, why can't I monitor the tasks I post on the GEE code editor? Whether it is through the Geemap function or the GEE task to start the download task, I have never been able to see any task generation in my GEE code editor. However, after a certain period of time, this task will indeed be completed in my Google Cloud Drive or my ASSET. Teacher Wu, do I have any way to check why this is happening? Or what advice does the teacher have? Where is the problem?

[jdbcode]

This is an Earth Engine API question, geemap is not involved. The 3rd paragraph here (https://developers.google.com/earth-engine/guides/processing_environments#task_lifecycle) describes the various ways that you can monitor tasks. Note that depending on Cloud Project settings you may not always be able to see tasks. Ensure that the project that you started the task from is the project that you are also using to check on the status of the task. If you think there is an issue, please file a bug (https://issuetracker.google.com/issues/new?component=184406&template=1254745).

[giswqs]

@jdbcode Good point. It might be caused the cloud project being used for exporting tasks is different from the one being used by the Code Editor. @zwy1502 Please check if you have multiple cloud projects under your account.

[zwy1502]

I tried running this code print (ee. data. getAssetRoots()) on the Python port, and the result was [{'type': 'Folder', 'id': 'projects/earthengine language/sets/users/zhouwang1502'}]. Does this mean that the Python side did not specify a cloud project? How should I specify it?

[zwy1502]

@jdbcode Good point. It might be caused the cloud project being used for exporting tasks is different from the one being used by the Code Editor. @zwy1502 Please check if you have multiple cloud projects under your account.
May I ask if Teacher Wu can teach me how to specify cloud projects for my Python client, or tell me how to view the specified projects for GEE Python client? I have confirmed that I only have one cloud project, and my method is to open the GEE code editor and display the name of my cloud project as ee my zwy in the upper right corner. Now, is it okay for me to just make sure that my GEE Python is also this project? How to view it? How to make changes to ensure consistency in the GEE code editor and Python?

[SelenaGeo]

I was having the same issue recently when re-running code that I previously ran using Geemap/the EE API about 5 months ago. A fix that worked for me is to state the project name when initialising: ee.Initialize(project='project_name')
Previously it seemed to just identify the default folder, so this wasn't always required.

An alternative work-around is to check the task status in Python:

tasks = ee.batch.Task.list()
for task in tasks:
    print(f'Task ID: {task.id}, State: {task.state}, Description: {task.config.get("description", "No description")}')

[giswqs]

Yes, a project ID is required now.