The output of softmax was overwritten by the output of fc2 in the following line. So, the output of the softmax is never utilized. Now, the final output of the model includes softmax.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/107988
Approved by: https://github.com/zou3519
This PR:
- dedups our colab notebooks with the regular functorch notebooks. The
colab notebooks were versions of the reuglar notebooks that had
install instructions. Now that functorch is easier to install, we do not
need those anymore.
- fixes the colab links
Test Plan:
- build docs locally and tested them
Pull Request resolved: https://github.com/pytorch/pytorch/pull/85855
Approved by: https://github.com/samdow
* Create readme.md
* colab version, jacobians hessians tutorial
* per sample gradients tutorial in colab
* colab ready tutorial, ensembling
* add colab badge and link to colab ready tutorial
* Delete ensembling_colab.ipynb
* add colab badge and link to colab ready version
* add colab badge and link to colab ready tutorial
* add colab badge and link to colab ready tutorial, add meta-learning mention as addtl use case