search

UDST / synthpop

Synthetic populations from census data
BSD 3-Clause "New" or "Revised" License
100 stars 47 forks source link

Synthesizer Refinements #21

Closed jiffyclub closed 9 years ago

jiffyclub commented 9 years ago

Add drawing of households and then people so we get both, and calculate a metric of comparison how our draws compare to the target constraints. So far not doing anything with that.

coveralls commented 9 years ago

Coverage Status

Coverage increased (+0.05%) when pulling 04ab97118655fe7df82b8e51588570078e6d9e1a on synth into 6e1c2dcc1dd8aec0f34c10d33f13a1fcb26191b1 on master.

coveralls commented 9 years ago

Coverage Status

Coverage increased (+0.05%) when pulling ad80a0d9fa21e4cd8661d908dc63e6e99b8254bf on synth into 6e1c2dcc1dd8aec0f34c10d33f13a1fcb26191b1 on master.

coveralls commented 9 years ago

Coverage Status

Coverage increased (+0.06%) when pulling f68e43a404f22934e420e6b7c48c0637e94f3e91 on synth into 6e1c2dcc1dd8aec0f34c10d33f13a1fcb26191b1 on master.

jiffyclub commented 9 years ago

Now synthesize_all returns households, people, and fit quality data. The fit quality data is a dictionary of namedtuples. Keys are the geography IDs as tuples: (state, county, tract, block_group). Values are tuples of (household_chisq, household_p, people_chisq, people_p). Both are actually namedtuples, so the values can also be accessed by name like qual.household_chisq or geog.state.