Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Avoid np.log of zero in ECM #203

Open
wants to merge 2 commits into
base: master
Choose a base branch
from
Open

Avoid np.log of zero in ECM #203

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
5af6048
Avoid np.log of zero
emuccino Feb 21, 2024
46a3eca
Merge pull request #1 from emuccino/emuccino-safe-log
emuccino Feb 21, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
8 changes: 5 additions & 3 deletions recordlinkage/algorithms/nb_sklearn.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,6 +35,8 @@

_ALPHA_MIN = 1e-10

def log(x):
return np.log(x + np.finfo(float).eps)

def check_is_fitted(estimator, attributes, msg=None, all_or_any=all):
"""Perform is_fitted validation for estimator.
Expand Down Expand Up @@ -595,10 +597,10 @@ def fit(self, X):
g_freq_sum = g_freq.sum(axis=0)

# maximisation step
class_log_prior_ = np.log(g_freq_sum) - np.log(X.shape[0]) # p
class_log_prior_ = log(g_freq_sum) - log(X.shape[0]) # p
feature_prob_ = safe_sparse_dot(g_freq.T, X_unique_bin)
feature_log_prob_ = np.log(feature_prob_)
feature_log_prob_ -= np.log(np.atleast_2d(g_freq_sum).T)
feature_log_prob_ = log(feature_prob_)
feature_log_prob_ -= log(np.atleast_2d(g_freq_sum).T)

# Stop iterating when the class prior and feature probs are close
# to the values in the to previous iteration (parameters starting
Expand Down