Eigendecompositions

https://d2l.ai/chapter_appendix-mathematics-for-deep-learning/eigendecomposition.html

Why tensor flow section has a completely different value for eigenvalues and eigenvectors than what is there in mxnet and pytorch section? It seems tensorflow is doing some other calculation for the same stuff. Can someone please explain?

I think there is a typo in the tutorial where tf.linalg.eigh is being used instead of tf.linalg.eig.

Hi @Pratik_Srivastav, great catch! Would you like to post a PR and contribute to our project?

Hi @goldpiggy. Thanks. Please let me know if there is a procedure to be followed for raising a PR for the d2l repo as i am very new to d2l tutorial and just catching up.

Hi @goldpiggy. I have raised a PR https://github.com/d2l-ai/d2l-en/pull/1502 for the same as per the rep contribution guidelines. Thanks

Amazing! Thanks for contributing to D2L! Please feel free to post PR or ask question in the future.