a cli for interacting with stumpf server
Project description
installation
# install pipx
python3 -m pip install --user pipx
python3 -m pipx ensurepath
# install boiler tools
pipx install diva-boiler
boiler login stumpf
usage
This documentation provides some useful examples, but is not exhaustive.
# to get up-to-date documentation, use the help option
boiler [noun] --help
boiler [noun] [verb] --help
local data validation
boiler has utilities to validate KW18 data. These are offline operations.
# show kw18 help
boiler kw18 validate --help
# validate a single set a kw18 files
boiler kw18 validate examples/kw18/minimal
# find and validate all kw18 files recursively
boiler kw18 validate -r examples/kw18
For this command, it is assumed that all files associated with a KW18
dataset have the same base name and there is only one *.kw18 file
per directory.
searching for videos
boiler video search --help
boiler video search --name 2018-03-15.10-35-00.10-40-00.school.G336
boiler video search --scenario basketball --gtag G336
By default, this command will return only the first 20 matches. You can get the next
set of results using the --page option.
boiler video search --page 2
downloading annotation data
The following command will download KPF files for one or more videos to the current directory. This will fetch the most recent version of all data as it currently exists in the system (including unaudited annotations).
boiler export video --help
boiler export video 2018-03-15.10-35-00.10-40-00.school.G336 2018-10-15.17-05-00.17-10-00.alb.G333
video pipeline
dispatch annotation tasks to vendors
Once a video exists, it can be transitioned to the annotation state.
boiler vendor dispatch --help
# generate the list of known activity types to file
boiler activity list-types > activity-list.txt
# specify a video, vendor, and list of activities to transition to the annotation stage
boiler vendor dispatch \
--name kitware \
--video-name 2999-01-01.00-00-00.00-05-00.admin.G999 \
--activity-type-list activity-list.txt \
--set-name test-set
--annotation-repo-path iMerit/2018-09-06/08/2018-09-06.08-20-00.08-25-00.alb.G229/37-activities
Alternatively, the dispatch can be done after the results have been returned from the vendor. This can be done in bulk as follows:
boiler kw18 dispatch \
--vendor imerit \
--set-name testing \
--activity-type-list activity-list.txt \
m2-annotations-imerit/iMerit/2018-09-06/*/37-activities
This command also support a --recursive flag to search recursively for kw18 files under each
listed path.
vendor activity ingestion
When activities come back from vendors, they should be transitioned to the audit state.
boiler kw18 ingest --help
# list each directory explicitly to ingest
boiler kw18 ingest m2-annotations-imerit/iMerit/2018-09-06/*/37-activities
# or run recursively
boiler kw18 ingest --recursive m2-annotations-imerit/iMerit
Stumpf will first detect whether the files have changed or not. If they have not, no further action will be taken. If they have, then Stumpf will:
- Generate a transition to the "annotation" status
- Run server side validation
- If validation fails, return failure information
- If validation succeeds, transition to the "audit" state and ingest activities from the KW18 files
audited annotation ingestion
To dispatch videos to gunrunner, use the following command.
boiler gunrunner dispatch --help
boiler gunrunner dispatch --activity-type-list 47-activities.txt \
m2-annotations-audit/2018-10-15/08/2018-10-15.08-15-00.08-20-00.alb.G333/47-activities
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
File details
Details for the file diva-boiler-0.1.28.dev16.tar.gz.
File metadata
- Download URL: diva-boiler-0.1.28.dev16.tar.gz
- Upload date:
- Size: 173.6 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/3.2.0 pkginfo/1.5.0.1 requests/2.24.0 setuptools/49.2.1 requests-toolbelt/0.9.1 tqdm/4.48.2 CPython/3.6.11
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 | 5b0cd024b97b4b3ebdc95c4a8580781b9e27998395419788e00b8d2ecedb8421 |
|
| MD5 | 61b1541824d062f37dc7e0b282abff02 |
|
| BLAKE2b-256 | aeef7b1451d07ecfcc0b5801c2afd1bd7db47a75631ac5d703c6edb518adda27 |
