BUGFIX: close pool for emcee saving

An external user pointed out that saving a result with emcee after using parallelization fails. This was due to a missing _close_pool which has now been added.

changed milestone to %2.2.0

added <10 lines Bug Sampling labels

added 1 commit

• 60d4385a - TEST: make sure sampler run test saves the result

Compare with previous version

marked this merge request as ready

changed title from Draft: Update emcee.py to BUGFIX: close pool for emcee saving

added 1 commit

• ead0a9e4 - BUGFIX: make dnest4 not store unhelpful loglikelihoods

Compare with previous version

added 1 commit

• b5caa6aa - TEST: dnest4 doesn't store log-likelihood evaluations

Compare with previous version

added 1 commit

• 7290ec94 - FORMAT: remove unused import

Compare with previous version

While fixing the bug in emcee, I realised that the log_likleihood evaluations stored for dnest4 analyses were not the correct quantity. I couldn't figure out how to extract the actual log-likelihood values so I just removed the clause where they were extracted.

approved this merge request

approved this merge request

Approved on the basis that only the Python 3.8 tests are failing, which will be removed in !1277 (merged)

mentioned in issue #694 (closed)

merged

mentioned in commit 6bb226d9