Skip to content

Fix transpose order in prior marginal variance formula in tutorial_linear_regression.ipynb - #824

Open
Phuocminh94 wants to merge 1 commit into
mml-book:masterfrom
Phuocminh94:patch-1
Open

Fix transpose order in prior marginal variance formula in tutorial_linear_regression.ipynb#824
Phuocminh94 wants to merge 1 commit into
mml-book:masterfrom
Phuocminh94:patch-1

Conversation

@Phuocminh94

Copy link
Copy Markdown

Fixes a linear algebra transpose typo in the markdown text of Section 3 (Bayesian Linear Regression).

The formula for the prior (marginal) variance of the latent function was written as an outer product ($\alpha^2 \boldsymbol{\phi}(\boldsymbol{x}*) \boldsymbol{\phi}(\boldsymbol{x})^\top$), which yields a matrix. This PR corrects the order to an inner product ($\alpha^2 \boldsymbol{\phi}(\boldsymbol{x}_)^\top \boldsymbol{\phi}(\boldsymbol{x}_*)$) to correctly evaluate to a scalar variance, matching the actual code implementation in the subsequent cell.

…near_regression.ipynb

Fixes a linear algebra transpose typo in the markdown text of Section 3 (Bayesian Linear Regression). 

The formula for the prior (marginal) variance of the latent function was written as an outer product ($\alpha^2 \boldsymbol{\phi}(\boldsymbol{x}_*) \boldsymbol{\phi}(\boldsymbol{x}_*)^\top$), which yields a matrix. This PR corrects the order to an inner product ($\alpha^2 \boldsymbol{\phi}(\boldsymbol{x}_*)^\top \boldsymbol{\phi}(\boldsymbol{x}_*)$) to correctly evaluate to a scalar variance, matching the actual code implementation in the subsequent cell.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant