Ert3 examples #1
Open
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Add first iteration of the API for the ensemble evaluator Co-authored-by: Frode Aarstad <[email protected]> Co-authored-by: Morten Bendiksen <[email protected]> Co-authored-by: Jonas Grønås Drange <[email protected]> Co-authored-by: Sondre Sortland <[email protected]>
Co-authored-by: Frode Aarstad <[email protected]> Co-authored-by: Jonas G. Drange <[email protected]> Co-authored-by: Sondre Sortland <[email protected]>
The Qt5 backend provides the same classes as Qt4, so it is a drop-in replacement. According to the documentation page below, 'FigureCanvas' is an alias of 'FigureCanvasQTAgg', so the import 'as' is unnecessary. https://doc.ebichu.cc/matplotlib/api/backend_qt5agg_api.html
Launch ert storager server if no server is found
This commit includes the following major changes: - Conversion from Flask to FastAPI - Removal of RdpApi and StorageApi classes, preferring to use SqlAlchemy without any additional wrappers. (It should be noted that the FastAPI cookiecutter project implements a bunch of CRUD functions that we could also implement. I decided not to because it's unnecessary at this point in time.) - Addition of POST-ing endpoints - Extraction now posts data via HTTP rather than directly via SqlAlchemy - "New Storage/Old Plotting" has been removed. It was a proof-of-concept which we intend to reimplement within the Webviz framework (webviz-ert)
When refactoring for FastAPI, the create_update function was incorrectly transferred. The lack of ensemble updates came from a bug where a function was inlined naively, causing an early return. The usage of the term 'misfit' is incorrect but I've chosen not to touch it. It should rather be called an update or a synonym thereof.
Starlette (and therefore FastAPI) explicitly disables float nans when using the `JSONResponse` object, which is the default one used by FastAPI. We replace the default with our own copy that has `allow_nan=True`.
This is a hack in order to avoid the latency from NFS. In the future this will be done properly by having a separate centralised database server to which ERT will connect to and thus we won't be on NFS anymore.
This PR introduces the use of Prefect as the workflow manager/queue system. The initial implementation can run both locally and on the Equinor LSF cluster. It uses a newly introduced configuration file to set up a Prefect ensemble that can run a UnixStep (a user-configured python script) as part of a forward model. Depending on the configuration of inputs and outputs of the UnixSteps that define the forward-model the steps can run either in parallel or sequentially. A UnixStep has the following assumptions: * Access to shared disk for all of the jobs * Data is loaded from the shared disk * Integration it with the current CLI/GUI * Supports only a programmatic interface. This first iteration has been tested using a poly case example. Resolves equinor#1125
Co-authored-by: Markus Dregi <[email protected]> Co-authored-by: Øyvind Eide <[email protected]> Co-authored-by: Anette Uttisrud <[email protected]> Co-authored-by: Lars Petter Øren Hauge <[email protected]> Co-authored-by: Helena Nandi Formentin <[email protected]> Co-authored-by: Sondre Sortland <[email protected]> Co-authored-by: anna.kvashchuk <[email protected]>
* Have ert3.engine stop closing passed streams A stream is passed to ert3.engine such that records can be loaded and persisted in storage. The stream is opened outside of the engine, but currently closed within. This is a bad design pattern and a remenant from when the engine code lived as part of the console submodule! Instead the one opening the stream should also be responsible for closing it. * Open files using context managers * Close record file explicitly after loading
Start clean ert-storage on test with fixture
kvashchuka
force-pushed
the
ert3_examples
branch
from
894c460 to
8811193
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Issue
Resolves #my_issue
Approach
Short description of the approach