Nicholas Vadivelu
Nicholas Vadivelu
The "New Libraries" section was meant to showcase cool libraries that are up-and-coming, to expose them to new users to help them develop and grow. However, some of the libraries...
[`prefetch_to_device`](https://flax.readthedocs.io/en/latest/flax.jax_utils.html#flax.jax_utils.prefetch_to_device) is great when doing distributed training, but can be a hassle for non-pmap use cases since you need to add then remove a dummy sharding dimension at the front....
Currently, `MaxPool` and `AvgPool` expect window shape and stride to be "Same rank as value or `int`.". I think the layers should check the window shape and size and ensure...
### Description The [sample function](https://github.com/SPFlow/SPFlow/blob/master/src/spn/experiments/RandomSPNs/RAT_SPN.py#L495-L507) for RAT_SPN says `# Does not work currently!`. What is wrong with it? It was introduced in this commit: https://github.com/SPFlow/SPFlow/commit/29282bbce21ff7abc3480aab93d482b4388e8bf1
The vectorized optimizer in [`mnist_dpsgd_tutorial_vectorized.py`](https://github.com/tensorflow/privacy/blob/master/tutorials/mnist_dpsgd_tutorial_vectorized.py) runs significantly faster than the new TF2 Keras optimizer used in [`mnist_dpsgd_tutorial_keras.py`](https://github.com/tensorflow/privacy/blob/master/tutorials/mnist_dpsgd_tutorial_keras.py). On a Titan RTX (24 GB VRAM) at batch size 250 and 250...
Thanks for this great linting tool. [These three items](https://github.com/n2cholas/awesome-jax/blame/5ee92181d73e506036e6ab46578f6daf686b8234/readme.md#L91-L93) produce the following error: ``` ✖ Linting readme.md:90:4 ✖ 90:4 Invalid list item link remark-lint:awesome-list-item ✖ 91:4 Invalid list item link...
As per @marcvanzee's comment on #1022, most of the traverse_util logic can be done with tree_maps without losing any run time optimizations. The trade-off is the compile times will be...
## What is this Python project? JAX is a machine learning library with a NumPy-like interface designed to run on accelerators such as GPUs and TPUs. Its distinguishing feature are...
**System information** - TensorFlow version (you are using): `2.5.0-dev20201115` - Are you willing to contribute it (Yes/No): No (don't know how to) **Describe the feature and the current behavior/state.** Currently,...