Fix epsilon according to dtype in LdaModel #1770
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
piskvorky
requested changes
Dec 7, 2017
gensim/models/ldamodel.py
Outdated
| @@ -286,6 +286,9 @@ def __init__(self, corpus=None, num_topics=100, id2word=None, | |||
| >>> lda = LdaModel(corpus, num_topics=50, alpha='auto', eval_every=5) # train asymmetric alpha from data | |||
|
|
|||
| """ | |||
| if dtype not in {np.float16, np.float32, np.float64}: | |||
| raise ValueError("Incorrect 'dtype', please choice one of numpy.float16, numpy.float32 or numpy.float64") | |||
gensim/models/ldamodel.py
Outdated
| @@ -498,7 +501,8 @@ def inference(self, chunk, collect_sstats=False): | |||
| # The optimal phi_{dwk} is proportional to expElogthetad_k * expElogbetad_w. | |||
| # phinorm is the normalizer. | |||
| # TODO treat zeros explicitly, instead of adding 1e-100? | |||
| phinorm = np.dot(expElogthetad, expElogbetad) + 1e-100 | |||
| eps = 1e-100 if self.dtype == np.float64 else (1e-35 if self.dtype == np.float32 else 1e-5) | |||
There was a problem hiding this comment.
Easier to read and maintain as a mapping?
dtype_to_eps = {
np.float64: 1e-100,
np.float32: 1e-35,
np.float16: 1e-5,
}And then also if dtype not in dtype_to_eps: ...
piskvorky
requested changes
Dec 7, 2017
gensim/models/ldamodel.py
Outdated
| @@ -498,7 +501,13 @@ def inference(self, chunk, collect_sstats=False): | |||
| # The optimal phi_{dwk} is proportional to expElogthetad_k * expElogbetad_w. | |||
| # phinorm is the normalizer. | |||
| # TODO treat zeros explicitly, instead of adding 1e-100? | |||
There was a problem hiding this comment.
Should update comment too: 1e-100 => epsilon.
gensim/models/ldamodel.py
Outdated
| @@ -498,7 +501,13 @@ def inference(self, chunk, collect_sstats=False): | |||
| # The optimal phi_{dwk} is proportional to expElogthetad_k * expElogbetad_w. | |||
| # phinorm is the normalizer. | |||
| # TODO treat zeros explicitly, instead of adding 1e-100? | |||
| phinorm = np.dot(expElogthetad, expElogbetad) + 1e-100 | |||
| dtype_to_eps = { | |||
There was a problem hiding this comment.
Better defined at module level, so __init__ can make use of the allowed keys? Now the same information is in three places, not DRY.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
That's small fix for #1656, big thanks @rmalouf for catch this potential bug #1656 (comment)
What's done:
numpy.float16,numpy.float32,numpy.float64)epsdepends ondtype(instead of hardcoded1e-100)CC: @piskvorky