FTPVL is a library for simplifying the data collection and visualization process for Symbiflow development. Although it was made with Symbiflow in mind, it is highly extensible for future integration with other software.
Take a look at the notebooks below to demonstrate the functionality of FTPVL.
- Using
HydraFetcher
and Processors - Styling tables with
SingleTableVisualizer
- Comparing two Evaluations using the internal dataframe
- Filtering and Aggregating an Evaluation
- Comparing multiple Evaluations
Extensive documentation, including a Getting Started guide, is available on ReadTheDocs.
Documentation of this library is generated in the docs/
folder by reading
the docstrings from the source code.
The website is generated using Sphinx
using the Read the Docs theme.
pip install -r requirements.txt
cd docs
make html
pandas
: for data management and processing (website)seaborn
: for colormap generation (website)jinja2
: for visualization generation (website)scipy
: for support of built-in aggregators(website)
requests-mock
: for mocking request object for testing fetchers (website)pylint
: for linting (website)pytest
: testrunner (website)coverage
: for coverage testing (website)sphinx
: for documentation generation (website)sphinx-rtd-theme
: for documentation generation (theme) (website)
- Add support for fetching using the absolute evaluation number instead of relative
- Add support for handling legacy Icebreaker processors
- Added evaluation concatenation and
CompareToFirst
processors for relative comparisons between evaluations. - Fixed
HydraFetcher
issues when fetching an older evaluation that contains more than one build artifact.
- Added support for filter and aggregator processors, fixes #9
- Added support for custom projects and jobsets in HydraFetcher.
- Added RelativeDiff processor.
- Updated some internal docstrings