Scikit-Learn Gaussian Mixture: How can log-probabilities be positive? [closed]
Multi tool use
Clash Royale CLAN TAG#URR8PPP
I am fitting a Gaussian Mixture model:gm = GaussianMixture(n_components=K)
gm.fit(features)
When I do:
gm.score_samples(features)
All of the scores, which are supposed to be: "weighted log probabilities for each sample." are positive.
Are they actually log-probabilities?
probability clustering python scikit-learn gaussian-mixture
asked Dec 20 '18 at 18:50
Eduardo G
364
closed as off-topic by Xi'an, kjetil b halvorsen, Peter Flom♦ Dec 21 '18 at 12:48
This question appears to be off-topic. The users who voted to close gave this specific reason:
- "This question appears to be off-topic because EITHER it is not about statistics, machine learning, data analysis, data mining, or data visualization, OR it focuses on programming, debugging, or performing routine operations within a statistical computing platform. If the latter, you could try the support links we maintain." – Xi'an, kjetil b halvorsen, Peter Flom
add a comment |
I am fitting a Gaussian Mixture model:gm = GaussianMixture(n_components=K)
gm.fit(features)
When I do:
gm.score_samples(features)
All of the scores, which are supposed to be: "weighted log probabilities for each sample." are positive.
Are they actually log-probabilities?
probability clustering python scikit-learn gaussian-mixture
asked Dec 20 '18 at 18:50
Eduardo G
364
closed as off-topic by Xi'an, kjetil b halvorsen, Peter Flom♦ Dec 21 '18 at 12:48
This question appears to be off-topic. The users who voted to close gave this specific reason:
- "This question appears to be off-topic because EITHER it is not about statistics, machine learning, data analysis, data mining, or data visualization, OR it focuses on programming, debugging, or performing routine operations within a statistical computing platform. If the latter, you could try the support links we maintain." – Xi'an, kjetil b halvorsen, Peter Flom
add a comment |
I am fitting a Gaussian Mixture model:gm = GaussianMixture(n_components=K)
gm.fit(features)
When I do:
gm.score_samples(features)
All of the scores, which are supposed to be: "weighted log probabilities for each sample." are positive.
Are they actually log-probabilities?
probability clustering python scikit-learn gaussian-mixture
asked Dec 20 '18 at 18:50
Eduardo G
364
I am fitting a Gaussian Mixture model:gm = GaussianMixture(n_components=K)
gm.fit(features)
When I do:
gm.score_samples(features)
All of the scores, which are supposed to be: "weighted log probabilities for each sample." are positive.
Are they actually log-probabilities?
probability clustering python scikit-learn gaussian-mixture
probability clustering python scikit-learn gaussian-mixture
asked Dec 20 '18 at 18:50
Eduardo G
364
asked Dec 20 '18 at 18:50
Eduardo G
364
asked Dec 20 '18 at 18:50
Eduardo G
364
asked Dec 20 '18 at 18:50
Eduardo G
364
asked Dec 20 '18 at 18:50
Eduardo G
364
364
closed as off-topic by Xi'an, kjetil b halvorsen, Peter Flom♦ Dec 21 '18 at 12:48
This question appears to be off-topic. The users who voted to close gave this specific reason:
- "This question appears to be off-topic because EITHER it is not about statistics, machine learning, data analysis, data mining, or data visualization, OR it focuses on programming, debugging, or performing routine operations within a statistical computing platform. If the latter, you could try the support links we maintain." – Xi'an, kjetil b halvorsen, Peter Flom
closed as off-topic by Xi'an, kjetil b halvorsen, Peter Flom♦ Dec 21 '18 at 12:48
This question appears to be off-topic. The users who voted to close gave this specific reason:
- "This question appears to be off-topic because EITHER it is not about statistics, machine learning, data analysis, data mining, or data visualization, OR it focuses on programming, debugging, or performing routine operations within a statistical computing platform. If the latter, you could try the support links we maintain." – Xi'an, kjetil b halvorsen, Peter Flom
add a comment |
add a comment |
1 Answer
1
active
oldest
votes
They supposedly are probability densities, not probabilities.
A probability density can be larger than 1, hence the log can be positive.
The documentation of sklearn should probably be fixed to reflect this.
answered Dec 20 '18 at 22:05
Anony-Mousse
29.6k54178
add a comment |
1 Answer
1
active
oldest
votes
1 Answer
1
active
oldest
votes
active
oldest
votes
active
oldest
votes
They supposedly are probability densities, not probabilities.
A probability density can be larger than 1, hence the log can be positive.
The documentation of sklearn should probably be fixed to reflect this.
answered Dec 20 '18 at 22:05
Anony-Mousse
29.6k54178
add a comment |
They supposedly are probability densities, not probabilities.
A probability density can be larger than 1, hence the log can be positive.
The documentation of sklearn should probably be fixed to reflect this.
answered Dec 20 '18 at 22:05
Anony-Mousse
29.6k54178
add a comment |
They supposedly are probability densities, not probabilities.
A probability density can be larger than 1, hence the log can be positive.
The documentation of sklearn should probably be fixed to reflect this.
answered Dec 20 '18 at 22:05
Anony-Mousse
29.6k54178
They supposedly are probability densities, not probabilities.
A probability density can be larger than 1, hence the log can be positive.
The documentation of sklearn should probably be fixed to reflect this.
answered Dec 20 '18 at 22:05
Anony-Mousse
29.6k54178
answered Dec 20 '18 at 22:05
Anony-Mousse
29.6k54178
answered Dec 20 '18 at 22:05
Anony-Mousse
29.6k54178
answered Dec 20 '18 at 22:05
Anony-Mousse
29.6k54178
29.6k54178
add a comment |
add a comment |