# Machine Learning

**URL:** https://discourse.julialang.org/c/domain/ml/24.md?page=56

[Latest](https://discourse.julialang.org/latest.md) · [Categories](https://discourse.julialang.org/categories.md) · [Tags](https://discourse.julialang.org/tags.md)

**Page:** 57

---

## [Julia end-to-end LSTM for one CPU](https://discourse.julialang.org/t/julia-end-to-end-lstm-for-one-cpu/3942)

<div class="topic-metadata">

**Author:** [@hpoit](https://discourse.julialang.org/u/hpoit)\
**Replies:** 24\
**Last updated:** [June 2, 2017, 10:51pm UTC](https://discourse.julialang.org/t/julia-end-to-end-lstm-for-one-cpu/3942 "2017-06-02T22:51:45Z")

</div>

I wonder how this core would look like in Julia Together with a Julia frontend How do I build a Julia abstraction for all of this, integrated with JuliaDB? The GPU part would not be a priority at the moment, as I …

---

## [Flux Updates](https://discourse.julialang.org/t/flux-updates/3533)

<div class="topic-metadata">

**Author:** [@MikeInnes](https://discourse.julialang.org/u/MikeInnes)\
**Replies:** 0\
**Last updated:** [May 5, 2017, 12:02pm UTC](https://discourse.julialang.org/t/flux-updates/3533 "2017-05-05T12:02:58Z")

</div>

I’ve just released Flux’s first major update since its release, which brings everything up to speed for 0.6 and nails down APIs for things like RNNs and training. I’ve also reworked the docs based on user feedback, so n…

---

## [Segmented\_sum in Knet](https://discourse.julialang.org/t/segmented-sum-in-knet/3396)

<div class="topic-metadata">

**Author:** [@Tomas\_Pevny](https://discourse.julialang.org/u/Tomas_Pevny)\
**Replies:** 0\
**Last updated:** [April 27, 2017, 5:55am UTC](https://discourse.julialang.org/t/segmented-sum-in-knet/3396 "2017-04-27T05:55:03Z")

</div>

Hello, I have started to play with Knet, but I am unable to reproduce my work related to multi-instance learning in it, since I need segmented\_sum operation from tensorflow. I have created a small example using Knet u…

---

## [ANN: MLDataPattern.jl](https://discourse.julialang.org/t/ann-mldatapattern-jl/3299)

<div class="topic-metadata">

**Author:** [@Evizero](https://discourse.julialang.org/u/Evizero)\
**Replies:** 6\
**Last updated:** [April 21, 2017, 1:32pm UTC](https://discourse.julialang.org/t/ann-mldatapattern-jl/3299 "2017-04-21T13:32:24Z")

</div>

Hello everyone! I am really happy to finally announce the next JuliaML package reaching a stable state: MLDataPattern.jl Github: GitHub - JuliaML/MLDataPattern.jl: Utility package for subsetting, resampling, iteration,…

---

## [Mocha Type assert failure](https://discourse.julialang.org/t/mocha-type-assert-failure/3189)

<div class="topic-metadata">

**Author:** [@dryden4482](https://discourse.julialang.org/u/dryden4482)\
**Replies:** 0\
**Last updated:** [April 12, 2017, 10:43pm UTC](https://discourse.julialang.org/t/mocha-type-assert-failure/3189 "2017-04-12T22:43:00Z")

</div>

Load error in loading a snapshot. The error is somewhat random. Sometimes it works, other times it does not. I believe the compose package is interfering with mocha. TypeError: read\_refs: in typeassert, expected Array{A…

---

## [t-SNE visualization](https://discourse.julialang.org/t/t-sne-visualization/3150)

<div class="topic-metadata">

**Author:** [@hpoit](https://discourse.julialang.org/u/hpoit)\
**Replies:** 0\
**Last updated:** [April 10, 2017, 9:25pm UTC](https://discourse.julialang.org/t/t-sne-visualization/3150 "2017-04-10T21:25:46Z")

</div>

Hi @malmaud, I’ve read all your files in TensorFlow.jl/docs/src/, thanks for that. Is there an advantage to integrating t-SNE to TensorFlow.jl? I’m not sure how I would use it separately either (haven’t studied it yet).…

---

## [Building LSTM](https://discourse.julialang.org/t/building-lstm/3153)

<div class="topic-metadata">

**Author:** [@hpoit](https://discourse.julialang.org/u/hpoit)\
**Replies:** 1\
**Last updated:** [April 11, 2017, 12:16pm UTC](https://discourse.julialang.org/t/building-lstm/3153 "2017-04-11T12:16:24Z")

</div>

How do I build the LSTM cell? Is it nn.rnn.LSTMCell(input, state, input\_dim=-1) ? Can I add that under RNN in https://github.com/malmaud/TensorFlow.jl/blob/master/docs/src/ops.md ? @malmaud

---

## [Dropout regularization](https://discourse.julialang.org/t/dropout-regularization/3154)

<div class="topic-metadata">

**Author:** [@hpoit](https://discourse.julialang.org/u/hpoit)\
**Replies:** 1\
**Last updated:** [April 11, 2017, 11:39am UTC](https://discourse.julialang.org/t/dropout-regularization/3154 "2017-04-11T11:39:00Z")

</div>

Hi Jon @malmaud, is nn.dropout(cell, Tensor(output\_keep\_prob)) only for multi-cell nn? If my LSTM has only one cell state, then is there a need for it?

---

## [Tensorflow.jl - Where is the wrapper of \`tensorflow::ops::Abs\`?](https://discourse.julialang.org/t/tensorflow-jl-where-is-the-wrapper-of-tensorflow-ops-abs/3095)

<div class="topic-metadata">

**Author:** [@hpoit](https://discourse.julialang.org/u/hpoit)\
**Replies:** 1\
**Last updated:** [April 6, 2017, 11:40pm UTC](https://discourse.julialang.org/t/tensorflow-jl-where-is-the-wrapper-of-tensorflow-ops-abs/3095 "2017-04-06T23:40:36Z")

</div>

Hi. From TensorFlow C++ API Reference | TensorFlow v2.9.1, module math\_ops, operator tensorflow::ops::Abs, where is the wrapper of Abs( const ::tensorflow::Scope & scope, ::tensorflow::Input x ) \`\`\` in https://gi…

---

## [Adding regularization in TensorFlow.jl](https://discourse.julialang.org/t/adding-regularization-in-tensorflow-jl/3006)

<div class="topic-metadata">

**Author:** [@Tomas\_Pevny](https://discourse.julialang.org/u/Tomas_Pevny)\
**Replies:** 0\
**Last updated:** [April 2, 2017, 6:16am UTC](https://discourse.julialang.org/t/adding-regularization-in-tensorflow-jl/3006 "2017-04-02T06:16:48Z")

</div>

Hello All, I am starting to play with TensorFlow.jl. I am thinking about how to add l2 regularization into all weights of all layers. Adding that during the definition stage seems to me to be a nuisance. I would like so…

---

## [Introducing Seep.jl, a tensorflow like library for julia](https://discourse.julialang.org/t/introducing-seep-jl-a-tensorflow-like-library-for-julia/1254)

<div class="topic-metadata">

**Author:** [@adambrewster](https://discourse.julialang.org/u/adambrewster)\
**Replies:** 15\
**Last updated:** [March 31, 2017, 10:46pm UTC](https://discourse.julialang.org/t/introducing-seep-jl-a-tensorflow-like-library-for-julia/1254 "2017-03-31T22:46:26Z")

</div>

Introducing Seep.jl, a tensorflow like library for julia, available at https://github.com/mit-ll/Seep.jl. I wrote this package over the last year to learn about modern neural network architectures. It still immature an…

---

## [Building and unit testing Protocol Buffer in Tensorflow.jl](https://discourse.julialang.org/t/building-and-unit-testing-protocol-buffer-in-tensorflow-jl/2958)

<div class="topic-metadata">

**Author:** [@hpoit](https://discourse.julialang.org/u/hpoit)\
**Replies:** 3\
**Last updated:** [March 30, 2017, 1:01pm UTC](https://discourse.julialang.org/t/building-and-unit-testing-protocol-buffer-in-tensorflow-jl/2958 "2017-03-30T13:01:45Z")

</div>

Hi @malmaud. What’s your reference for building and testing protobuf?

---

## [TensorFlow.jl basic usage ERROR: UndefVarError: global\_variables\_initializer not defined](https://discourse.julialang.org/t/tensorflow-jl-basic-usage-error-undefvarerror-global-variables-initializer-not-defined/2951)

<div class="topic-metadata">

**Author:** [@hpoit](https://discourse.julialang.org/u/hpoit)\
**Replies:** 1\
**Last updated:** [March 29, 2017, 3:43pm UTC](https://discourse.julialang.org/t/tensorflow-jl-basic-usage-error-undefvarerror-global-variables-initializer-not-defined/2951 "2017-03-29T15:43:23Z")

</div>

julia\> using TensorFlow INFO: Precompiling module TensorFlow. julia\> sess=TensorFlow.Session() 2017-03-29 12:11:27.483376: W tensorflow/core/platform/cpu\_feature\_guard.cc:45\] The TensorFlow library wasn't compiled to us…

---

## [Interested in Summer of Code](https://discourse.julialang.org/t/interested-in-summer-of-code/2337)

<div class="topic-metadata">

**Author:** [@americast](https://discourse.julialang.org/u/americast)\
**Replies:** 6\
**Last updated:** [March 18, 2017, 11:59am UTC](https://discourse.julialang.org/t/interested-in-summer-of-code/2337 "2017-03-18T11:59:48Z")

</div>

Hello everyone, I am interested in applying for Google Summer of Code as a student this year under Julia. I would like to work on building a package similar to BinDeps, which would enable us to create data providing pac…

---

## [Dropout in mxnet](https://discourse.julialang.org/t/dropout-in-mxnet/2667)

<div class="topic-metadata">

**Author:** [@Patrik\_Waldmann](https://discourse.julialang.org/u/Patrik_Waldmann)\
**Replies:** 0\
**Last updated:** [March 14, 2017, 7:31pm UTC](https://discourse.julialang.org/t/dropout-in-mxnet/2667 "2017-03-14T19:31:10Z")

</div>

I’m trying to understand how to use dropout in mxnet and start with a simple regression example: #= Regression with dropout =# using MXNet # create training and evaluation data sets url = "https://archive.ics.uci.edu/m…

---

## [Julia implementations of some of the foundational Machine Learning models and algorithms from scratch](https://discourse.julialang.org/t/julia-implementations-of-some-of-the-foundational-machine-learning-models-and-algorithms-from-scratch/2586)

<div class="topic-metadata">

**Author:** [@memoiry](https://discourse.julialang.org/u/memoiry)\
**Replies:** 0\
**Last updated:** [March 10, 2017, 4:47am UTC](https://discourse.julialang.org/t/julia-implementations-of-some-of-the-foundational-machine-learning-models-and-algorithms-from-scratch/2586 "2017-03-10T04:47:54Z")

</div>

Hi I have written a project https://github.com/memoiry/lightML.jl including basic implementation of typical machine learning algorithms implemented in Julia. It’s a self-educational project so that the implementation is…

---

## [Using Neural Network for regression (in Julia)](https://discourse.julialang.org/t/using-neural-network-for-regression-in-julia/2410)

<div class="topic-metadata">

**Author:** [@Balinus](https://discourse.julialang.org/u/Balinus)\
**Replies:** 11\
**Last updated:** [March 6, 2017, 4:01pm UTC](https://discourse.julialang.org/t/using-neural-network-for-regression-in-julia/2410 "2017-03-06T16:01:03Z")

</div>

Hello, I’d like to use Julia for a small ML project. My dataset is about 2000 samples and 142 features. I was hoping to use a simple Neural Network to do this task (with a ReLU neuron as the output node?). However, I j…

---

## [Flux: Machine Learning with Julia](https://discourse.julialang.org/t/flux-machine-learning-with-julia/2382)

<div class="topic-metadata">

**Author:** [@MikeInnes](https://discourse.julialang.org/u/MikeInnes)\
**Replies:** 8\
**Last updated:** [March 3, 2017, 5:58pm UTC](https://discourse.julialang.org/t/flux-machine-learning-with-julia/2382 "2017-03-03T17:58:23Z")

</div>

I’m pleased to announce the early alpha release of Flux, a Julia interface for machine learning. Flux gives you the best of both worlds. Like Knet or PyTorch, Flux code is easy to reason about; it behaves like Julia, yo…

---

## [Unable to understand char-lstm example in MXNET for Julia](https://discourse.julialang.org/t/unable-to-understand-char-lstm-example-in-mxnet-for-julia/2186)

<div class="topic-metadata">

**Author:** [@kishoreabhishek](https://discourse.julialang.org/u/kishoreabhishek)\
**Replies:** 0\
**Last updated:** [February 20, 2017, 1:53pm UTC](https://discourse.julialang.org/t/unable-to-understand-char-lstm-example-in-mxnet-for-julia/2186 "2017-02-20T13:53:30Z")

</div>

I am trying to understand the char lstm example mentioned here - https://github.com/dmlc/MXNet.jl/blob/master/examples/char-lstm/lstm.jl Function lstm\_cell accepts the second parameter as previous state - function lstm…

---

## [ANN: LossFunctions.jl](https://discourse.julialang.org/t/ann-lossfunctions-jl/2043)

<div class="topic-metadata">

**Author:** [@Evizero](https://discourse.julialang.org/u/Evizero)\
**Replies:** 3\
**Last updated:** [February 17, 2017, 2:14am UTC](https://discourse.julialang.org/t/ann-lossfunctions-jl/2043 "2017-02-17T02:14:01Z")

</div>

Hello everyone! After MLLabelUtils.jl, I am happy to announce the next JuliaML package to reach a mature state: LossFunctions.jl Github: GitHub - JuliaML/LossFunctions.jl: Julia package of loss functions for machine le…

---

## [Speed of Knet](https://discourse.julialang.org/t/speed-of-knet/1927)

<div class="topic-metadata">

**Author:** [@Tomas\_Pevny](https://discourse.julialang.org/u/Tomas_Pevny)\
**Replies:** 0\
**Last updated:** [February 5, 2017, 7:58pm UTC](https://discourse.julialang.org/t/speed-of-knet/1927 "2017-02-05T19:58:46Z")

</div>

Hello All, I would like to ask if anyone experienced with KNet can explain me, what I am doing wrong and why a simple code like this is incredibly slow push!(LOAD\_PATH,joinpath(homedir(),“Work/julia/Knet”)) using Pat…

---

## [Training LSTM in MXNet.jl with differing input and output?](https://discourse.julialang.org/t/training-lstm-in-mxnet-jl-with-differing-input-and-output/1707)

<div class="topic-metadata">

**Author:** [@TravisA9](https://discourse.julialang.org/u/TravisA9)\
**Replies:** 2\
**Last updated:** [January 30, 2017, 4:49pm UTC](https://discourse.julialang.org/t/training-lstm-in-mxnet-jl-with-differing-input-and-output/1707 "2017-01-30T16:49:35Z")

</div>

Sorry guys, I am new to this and don’t have good access to any courses or good books on the subject. My question could probably best be illustrated with a translator… the expected output would be quite different from the…

---

## [Queries regarding Symbolic API in MXNet in Julia](https://discourse.julialang.org/t/queries-regarding-symbolic-api-in-mxnet-in-julia/1753)

<div class="topic-metadata">

**Author:** [@kishoreabhishek](https://discourse.julialang.org/u/kishoreabhishek)\
**Replies:** 1\
**Last updated:** [January 29, 2017, 9:30am UTC](https://discourse.julialang.org/t/queries-regarding-symbolic-api-in-mxnet-in-julia/1753 "2017-01-29T09:30:40Z")

</div>

I am struggling to understand the symbolic API nuances of MXNet in Julia. I saw an example in MXNet documentation which has the following line: act1 = mx.Activation(data = fc1, name=:relu1, act\_type=:relu) Why is act\_ty…

---

## [Any Julia deep learning frameworks use automatic differentiation?](https://discourse.julialang.org/t/any-julia-deep-learning-frameworks-use-automatic-differentiation/1596)

<div class="topic-metadata">

**Author:** [@Phil\_Tomson](https://discourse.julialang.org/u/Phil_Tomson)\
**Replies:** 10\
**Last updated:** [January 24, 2017, 9:06pm UTC](https://discourse.julialang.org/t/any-julia-deep-learning-frameworks-use-automatic-differentiation/1596 "2017-01-24T21:06:48Z")

</div>

Do any of the current Julia-based deep-learning frameworks use automatic differentiation? I’m pretty sure Mocha.jl does not. MXNET.jl seems to, but it actually is just a wrapper around the mxnet C++ code (so the AD is ta…

---

## [ANN: Espresso.jl - an expression transformation package](https://discourse.julialang.org/t/ann-espresso-jl-an-expression-transformation-package/1641)

<div class="topic-metadata">

**Author:** [@dfdx](https://discourse.julialang.org/u/dfdx)\
**Replies:** 1\
**Last updated:** [January 22, 2017, 7:49pm UTC](https://discourse.julialang.org/t/ann-espresso-jl-an-expression-transformation-package/1641 "2017-01-22T19:49:34Z")

</div>

Espresso.jl is a new package for expression matching and transformation. A few functions that might be intersting for people working with Julia AST: matchex - match expression by pattern, extract matching elements; s…

---

## [Factor multiplication values: incorrect reshape?](https://discourse.julialang.org/t/factor-multiplication-values-incorrect-reshape/1394)

<div class="topic-metadata">

**Author:** [@hpoit](https://discourse.julialang.org/u/hpoit)\
**Replies:** 4\
**Last updated:** [January 11, 2017, 4:17pm UTC](https://discourse.julialang.org/t/factor-multiplication-values-incorrect-reshape/1394 "2017-01-11T16:17:24Z")

</div>

I can’t get Cvaluespace to output the correct values, identical to table 3. I think it has to do with reshape, notebook is below. https://github.com/hpoit/DFactor.jl/blob/master/function%20DFmultiply.ipynb Thanks …

---

## [ANN: MLLabelUtils.jl](https://discourse.julialang.org/t/ann-mllabelutils-jl/1233)

<div class="topic-metadata">

**Author:** [@Evizero](https://discourse.julialang.org/u/Evizero)\
**Replies:** 0\
**Last updated:** [January 1, 2017, 2:33am UTC](https://discourse.julialang.org/t/ann-mllabelutils-jl/1233 "2017-01-01T02:33:11Z")

</div>

Hello everyone and a happy new year! I am happy to announce a new JuliaML package for working with classification targets in various formats. It is not yet registered (so don’t believe the documentation!), but I will cr…

---

## [BoltzmannMachines.jl](https://discourse.julialang.org/t/boltzmannmachines-jl/1060)

<div class="topic-metadata">

**Author:** [@binderh](https://discourse.julialang.org/u/binderh)\
**Replies:** 8\
**Last updated:** [December 21, 2016, 2:29pm UTC](https://discourse.julialang.org/t/boltzmannmachines-jl/1060 "2016-12-21T14:29:01Z")

</div>

Hi everybody, We just put our package BoltzmannMachines.jl online. It provides a specific type of Deep Learning approach, which isn’t covered by typical Deep Learning packages, but nevertheless rather successful in appl…

---

## [Salsa machine learning test](https://discourse.julialang.org/t/salsa-machine-learning-test/1085)

<div class="topic-metadata">

**Author:** [@Mohan\_Rao](https://discourse.julialang.org/u/Mohan_Rao)\
**Replies:** 0\
**Last updated:** [December 21, 2016, 8:43am UTC](https://discourse.julialang.org/t/salsa-machine-learning-test/1085 "2016-12-21T08:43:42Z")

</div>

Hi, I am interested in ML using SALSA and ran it on iris data successfully. However, how do I interpret its output and do a test case. If it learnt the lengths of sepals & petals for each of classes then how do I inpu…

---

## [What beats a flat array with strides for variables?](https://discourse.julialang.org/t/what-beats-a-flat-array-with-strides-for-variables/751)

<div class="topic-metadata">

**Author:** [@hpoit](https://discourse.julialang.org/u/hpoit)\
**Replies:** 4\
**Last updated:** [December 6, 2016, 4:53pm UTC](https://discourse.julialang.org/t/what-beats-a-flat-array-with-strides-for-variables/751 "2016-12-06T16:53:53Z")

</div>

Context - 10.A.1 I am wondering on how to support 10.A.1 on a range of data representations, managed by a dispatch system, for a library of operations (product, marginalization, maximization, reduction, …) I don’t want…

[Previous page](https://discourse.julialang.org/c/domain/ml/24.md?page=55)

[Next page](https://discourse.julialang.org/c/domain/ml/24.md?page=57)