🚀 Feature

Add numerically stable cumulative logsumexp function. Also we have associated PR on cummax that is needed for numerically stable implementation (pytorch/pytorch#20240).

Motivation

This is useful when computing sum of probabilities and have different applications.

Pitch

Torch has cumsum and cumprod so I suggest logcumsumexp to be added.

Currently Alacritty allows sending arrow keys in the alternate screen buffer through the scrolling.faux_multiplier configuration option. However, there is a terminal escape sequence dedicated to set and unset this option.

According to the XTerm documentation CSI ? 1007 h should enable sending arrow keys in the alternate screen mode and CSI ? 1007 l should disable it.

XTerm by default h

Issue Description

I'm unable to run DL4J with nvidia CUDA back end despiite following the instructions here:
https://deeplearning4j.org/docs/latest/deeplearning4j-config-gpu-cpu
Project works fine with native back end. When I debug, I can see the service loader finding the JCublasBackend.java class and then failing on isAvailable().

As far as I can tell I've done everything recommende

The problem as far as I understand is that libfoo.so's only knowledge about libcuda.so is its SONAME, which is libcuda.so.1, so a binary wanting to link in libfoo.so needs to know who to find libcuda.so.1 in order to verify the available symbols. Generally, ld's suggestion is to use -rpath-link to provide the path to the library, but the nvidia docker repository doesn't come with the symlink to li

Hi,

Is there a plan to support categorical features when exporting to Onnx format?

Hi all,
So I'm using DIGITS version: 6.1.1 with Caffe version: 0.17.0 and Caffe flavor: NVIDIA
I've successfully used NVIDIA/Digits many times but when using compatible mode I cant replicate results in opencv or caffe BVLC.

This is the output I get on Digits:

but this is what I

is it Grid Search can solve CASH problems with NNI , it seems that it is usually used for hyper-parameters optimization, have you guys have finished some revision for Grid Search for solving CASH problems.

about Cash problems can refer to ：microsoft/nni#1178

We would like to forward a particular 'key' column which is part of the features to appear alongside the predictions - this is to be able to identify to which set of features a particular prediction belongs to. Here is an example of predictions output using the tensorflow.contrib.estimator.multi_class_head:

{"classes": ["0", "1", "2", "3", "4", "5", "6", "7", "8", "9"],
 "scores": [0.068196

Current implementation of join can be improved by performing the operation in a single call to the backend kernel instead of multiple calls.

This is a fairly easy kernel and may be a good issue for someone getting to know CUDA/ArrayFire internals. Ping me if you want additional info.

Hey everyone!

mapd-core-cpu is already available on conda-forge (https://anaconda.org/conda-forge/mapd-core-cpu)

now we should add some instructions on the documentation.

at this moment it is available for linux and we are working to finish that for osx.

some additional information about the configuration:

1. for now, always install omnisci-core-cpu inside a conda environment (als

Is your feature request related to a problem? Please describe.
In pandas, the _constructor_sliced property of a dataframe will return a pandas-Series constructor. I would like the same property to exist for a cudf dataframe. This would allow cudf series objects to be generated

This is the placeholder for information regarding the building of deepdetect on Ubuntu 16.04 LTS (expected future reference platform).

Build status: successful, tested with Caffe back-end on CPU

Thanks to @MartinThoma the correct way of doing it is below:

$ sudo apt-get remove libcurlpp0
$ cd [wherever]
$ git clone https://github.com/jpbarrette/curlpp.git
$ cd curlpp
$ cmake .
$ s

Could the documentation describe the output? Or maybe a heading row in the output. In particular, the second column, the percentage value.