Mohammed Ridwanul
Mohammed Ridwanul
**Is your feature request related to a problem? Please describe.** When you visit the dashboard, you are most likely to look for the job you just submitted. Similarly all queued...
**Is your feature request related to a problem? Please describe.** A user may tag multiple jobs with the name of a particular architecture. They should be able to filter all...
Find the utility and helper packages (two are within contrib) and merge them into a single place for this type of generic, useful code that is not really domain specific....
We have faced this issue before, but for some reason it was ignored. Recently it was resurfaced by https://github.com/SebastianGergen again in the Slack community channels. Doing `log_metric()` with anything other...
Currently if you go into a project and click on send to Tesorboard, it will create a server where it will run Tensorboard for that specific job. However this is...
As a user I should be able to go to the GUI, click on create a notebook, specify the amount of GPU and RAM the notebook should use. If those...
Recently we have had a request to use an already existing KeyCloak authentication server. With the way Atlas currently stand - is there any way for us to be able...
I'm going to log flaky tests in this ticket . They waste our time by having to rerun the pipeline. - [ ] test_put_has_status_code_and_params - [x] test_multiple_keys_with_at_most_singly_nested_values - [ ]...
Currently, the frontend does some heavy processing (`O(n^2)`) to convert the job data from the backend to the format that the chart library expects. This leads to long load times...
We currently support a plugin based architecture to be able to use both the local-docker-scheduler and the old K8s scheduler. We should remove this and migrate to the local-docker-scheduler moving...