* Improve diffexp for tiledb
- The rows from the A and B sets are gathered and processed at the same time. In this
way the matrix is only accessed once instead of twice for each tile.
- There is now a single thread queue that gets shared between all callers of the diffexp.
This will slow down work if diffexp gets too busy.
- There is a target_workunit amount of work given to each thread. Previously the
workunit was (rows selected * width of tile), which could be small. Now multiple
column tiles can be combined into one workunit. If the target is too small then
thread and other overheads may reduce performance. If target_workunit is too large
then the size of the gathered sub matrix may take up too much memory.
- add configuration parameters (max_workers, cpu_multiplier, and target_workunit)
* Specialize diffexp for tiledb
This patch adds a new diffexp algorithm which is tuned for tiledb.
This algorithm was written by Bruce and is adapted here to plug into the
current framework. The anndata_adaptor still calls the original
algotithm (which was move from diffexp.py to diffexp_generic.py).
The cxg_adaptor now calls the new diffexp_tiledb version. Some
code is shared between the two.
This is part 1 of the diffexp for tiledb. Further tuning and
global throttles are still needed.
A script to run and time diffexp with various options is also
added: test/run_diffexp.py.
* s3 region should have a single config param
The s3 region can also now be automatically determined to further
reduce errors.
This patch also fixes a bug with order of handling the config params.
The tiledb config needs to be fixed before attempting to load
(need to handle_adaptor before handle_single_dataset)
* warning on maxCount for diffexp
* cleanup logging
* clarification
* make the limits configurable
* make diff exp limit work
* danger!
* remove debugging code
* fix merge with master
* fix unit tests
Co-authored-by: Colin Megill <colinmegill@gmail.com>
fixes an issue with "cellxgene launch" which had a bad interaction between
command line parameters and config file parameters.
Now, the config files are applied first, followed by the parameters that
were provided in the command line.
There is also now a check that each of the config attributes is type checked.
* Improvements to the matrix cache
- Add a timelimit for the matrix in the cache.
Once the timelimit is reached, the matrix can be removed.
- If a DatasetAccessError occurs, then remove the dataset
from the matrix cache.
Fixes#1322
There is a small chicken and egg problem.
The config file could be in s3, therefore when using the DataLocator to
download the config file, we don't yet have an app_config object.
Adding a check to handle this case.
Mostly this is just instructions for how to do this,
with a small addition to the makefile.
This enables support for serving the about_legal_tos and about_legal_privacy
from the cellxgene server.
* Added a config hook for secret key into the app.
the server first looks in an environment variable,
then looks in a config file.
For the cellxgene launch app, a default key is used if none is provided.
For the eb app, a secret key must be provided.
See sample of current behavior below:
```
venv❯ cellxgene prepare example-dataset/pbmc3k.h5ad
[cellxgene] Starting CLI...
Error: [cellxgene] cellxgene prepare has not been installed. Please run
`pip install cellxgene[prepare]` to install the necessary requirements.
~/workspace/cellxgene mweiden/446-custom-color-palette*
1 venv❯ pip install cellxgene[prepare]
zsh: no matches found: cellxgene[prepare]
```
Fix:
Wrap cellxgene[prepare] in single quotes.
* Improved fix for matrix cache handling.
During the MatrixDataCacheItem acquire function there was a
time when the write lock was released and the read lock was taken.
During that time, the dataset could have been deleted, later
result in the MatrixDataCacheManageri data adaptor returning None.
The solution is to demote the writer lock to a reader lock instead
of unlocking and relocking.
Also, when a the cache needs to delete an entry, the delete
is done outside the MatrixDataCacheManager lock. This operation
only requires the write lock for the MatrixDataCacheItem.
Fixes#1255
* work around anndata bug 344
* fix accidental cut and paste error
* Use modified make_index_unique function
Temporarily copy code from https://github.com/theislab/anndata/pull/345
until the issue is resolved and released.
* Add notes and test for make_index_unique
* Lint fix
* Format python
Co-authored-by: Matt Weiden <538456+mweiden@users.noreply.github.com>
* black reformat
* tighten up error reporting
* lint
* fine tuning
* additional improvements in exception handling
* lint
* include exception and traceback in log
* fix typo
* Introduce a config file to cellxgene
The config file format is in yaml. The default config is located
in server/common/default_config.py. A user may create a yaml file
that contains a subset of these fields. It can be used during cellxgene
launch, or for hosted cellxgene.
The code has also been refactored. Much of the logic to check arguments
has moved from launch to app config.
It is now possible to set the tiledb context parameters using the config
file. Other feature will soon be handled in a similar way.
* Improve hosted cellxgene
- option to turn off the test index page, or supply a page for redirect.
For EB, The default is to return 404. For cli launch, the default is the test page.
- option to select which matrix types are allowed for multi dataset servers.
For EB, The default is CXG only. For cli launch, the default is any matrix type.
- Return early with an error response if diffexp is requested when not configured
- Verified that reembedings and user annotations also return with an error response
if used when not enabled.
TODO: The new options cannot currently be set by the user.
I plan to add a configuration file where these and all other settings can be set.
Fixes#1210Fixes#1228Fixes#1229
* Fix for favicon with --dataroot
* fix static assets in hosted cxg
The web proxy at aws eb was not finding the static assets.
The solution here is very simple: just copy the directory
containing the static assets to the top level of the artifact.zip.
This is not really the ideal solution. According to the AWS
docs you can make a mapping to the correct location in an
an ebextentions config file. I tried this and many combinations but
was not able to get this to work following that pattern.
Since we control the construction of the zip file, the solution
here isn't bad, but it could probably be made better.
