Metadata-Version: 2.1
Name: wf-smcmodel
Version: 0.0.1
Summary: Provide estimation and simulation capabilities for sequential Monte Carlo (SMC) models
Home-page: https://github.com/WildflowerSchools/wf-smcmodel
Author: Theodore Quinn
Author-email: ted.quinn@wildflowerschools.org
License: UNKNOWN
Keywords: Bayes,SMC
Platform: UNKNOWN
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python
Requires-Dist: wf-datetime-conversion (>=0.0.1)
Requires-Dist: tensorflow (>=2.2)
Requires-Dist: numpy (>=1.19.0)
Requires-Dist: python-dateutil (>=2.8.1)
Requires-Dist: tqdm (>=4.46.1)

# smcmodel

Provide estimation and simulation capabilities for sequential Monte Carlo (SMC) models

## Task list

* Add logging
* Add docstrings
* Generate documentation
* Add explicit names to tensors and ops
* Check dataflow graphs using tensorboard (are all objects of known size at graph specification time)
* In simulate, do init and pull initial values in one run
* Figure out how to force assignment ops without fetching their outputs
* Consider adding num_samples argument to simulate()


