Questions tagged [neural-networks]
Questions involving defining, training, executing, importing and exporting neural networks.
450
questions
0
votes
1answer
48 views
Custom function in Neural Network Layer
I am trying to create a custom function in a Neural Network. The output of the NN is a n x 2 matrix which represent coordinates of points in a plane. From this, the Euclidean pairwise distance matrix ...
0
votes
0answers
30 views
Data set for Neural Network
I am using Regression with Neural Networks and I want to set my train and test dataset.
But I have a huge dataset in Excel that obviously do not have the shape that Mathematica requires.
If I ...
3
votes
1answer
190 views
How to simulate a cellular automaton on a network?
Unfortunately, I have not yet understood the Neural Network well.
Suppose there is simple network (like Kauffman automata):
several nodes that can only be in states 1-0, connections between them, and ...
3
votes
1answer
60 views
Visualisation of edges values on weight network
I have a problem when I want to see
values of edges on my weight graph.
...
2
votes
1answer
99 views
Neural Network Simple task
I'm trying to make a simple neural network to be able to find the y valeu given x1, x2 and x3. So first I made the training data:
...
1
vote
0answers
55 views
Create New Type of Layer For Neural Network
I'm doing some work exploring modelling neural networks in which each neuron could represent a different type of physical system. What I want to do is, I think, quite different to what an ordinary ...
3
votes
1answer
79 views
Save neural network after training
I've written quite a few projects that involve training a neural network through the Classify and Predict functions, but every ...
3
votes
3answers
191 views
Atypical graph partition
Suppose we have the given network 'graph', which we get based on the set 'edges':
...
1
vote
0answers
37 views
Importing A Neural Network From Mathematica For Use In R
I am experimenting with platform interoperability between Mathematica and R.
My aim is to create an untrained Neural Network using Mathematica, export this network in MXNet format as a .json file, ...
3
votes
0answers
56 views
How to efficiently implement a sub-pixel convolution layer in a CNN for image super-resolution
I am playing around using convolution neural networks for image super-resolution, i.e. up-scaling of images. In literature I often come across the concept of sub-pixel convolution as one of the ...
1
vote
0answers
25 views
How to define more general boundary conditions for ConvolutionLayer?
In Mathematica 12, I only see the option to define "PaddingSize" for a ConvolutionLayer in order to get Dirichlet boundary ...
2
votes
0answers
46 views
Are there any examples of pruning Neural Networks in Mathematica?
Are there any examples of Pruning in Neural Networks in Mathematica?
Any links or insight would be appreciated.
Thank You
Michel
0
votes
0answers
48 views
1
vote
0answers
40 views
Using LinearLayer for NetEncode[{“Characters”}]
So, I'm trying to make a net to translate a phrase. But I'm uncertain of the use of LinearLayer for a NetEncoder for Characters and getting some error in the training process. Here a toy example of ...
0
votes
0answers
47 views
Interpolated Hyperbolic tangent as a layer of a neural network
It is necessary to specify piecewise linear interpolation of the hyperbolic tangent. And substitute this function as a layer of a neural network. I tried a lot of things, please suggest your options.)
...
0
votes
0answers
56 views
Network training error InvalidJSON
I want to insert piecewise linear interpolation of the hyperbolic tangent into the neural network. An error occurs at the stage of training the network.
trainingData:
...
1
vote
2answers
98 views
Adding an interpolated function to neural networks through ElementwiseLayer
1.Trying to add an interpolated function to neural networks.
...
2
votes
1answer
65 views
Creating Loss Ports For Multiple Output Neural Net
I am making a multi-classfication neural net for a set of data. I have created the net but i think i need to specify a loss port at for each classification
Here are the labels for the classification ...
4
votes
0answers
66 views
Seq2Seq RNN with an AttentionLayer
In many Sequence to Sequence machine learning tasks, an Attention Mechanism is incorporated.
Before Transformer Networks, introduced in the paper: Attention Is All You Need, mainly RNNs were used to ...
4
votes
0answers
30 views
Appending a method to the Classify Function
Recently i created a color scheme and appended this to the relevant DataPaclet
...
0
votes
0answers
37 views
How can I train a neural net on Mathematica to predict text output from input?
I want to feed a neural net examples like this:
Input: I w e n t to the store
Output: I went to the store
Input: I was p l a y i n g with my friends
Output: I was playing with my friends
etc. ...
3
votes
1answer
157 views
How to paraphrase text?
I'm looking to train or use pre-trained models in order to re-phrase text. Are there any natural language models in the Wolfram Neural Repository (or elsewhere) that can be used to paraphrase text?
...
3
votes
0answers
28 views
PoolingLayer[] doesn't export to cloud anymore
Since last week (?) PoolingLayer[] appears broken when exported to the cloud from my local machine.
Lets define two nets, one is a poolingLayer and the other a ...
3
votes
0answers
53 views
Simple neural network with feedbacks in state-space form
Having studied more or less in detail the mechanisms of working with the state-space that Mathematica provides us with, I asked myself: can these tools be applied to neural networks?
This article ...
6
votes
1answer
75 views
Possible bugs in NetTrain? The loss of validation set shouldn't be lower than that of training set
I'm very curious why the loss of validation set is even lower than the loss of training set when I use NetTrain.
Say, on this page, https://reference.wolfram.com/language/tutorial/...
2
votes
1answer
910 views
GPU support on latest MacBook Pro, especially for deep learning
I'm thinking about getting some new top of the line computer hardware, specifically a MacBook Pro from Apple. These have AMD Radeon Pro 5500M GPUs, which while a bit slower than something like an ...
2
votes
1answer
30 views
How to change Input Port dimensions in NetChain containing NetMapOperator
Today I tried to build a fully convolutional 3D DeconvolutionLayer. However I encountered a peculiar problem.
Lets define the 3D layer as:
...
3
votes
0answers
44 views
Transforming a checkerboard using Transpose and ReshapeLayers
In the paper for RealNVP, they introduce a "squeezing" operation that takes a matrix and transforms it to have more channels, according to the following transformation (before and after):
As you can ...
1
vote
0answers
32 views
Loss layer for vectors with component mixing or non-trainable number array in NN layer
Suppose we need to construct an unconventional loss function for my neural network: vector MSE but taking into account the "similarity" of variables.
My network takes data vectors ($m=40$ D) and ...
5
votes
1answer
128 views
How to simplify this NetGraph a bit?
I have a NetGraph that takes list of sublists, and outputs a real value that is higher when there are more near duplicate sublists.
For example, run this code ...
1
vote
1answer
61 views
How to train a deep NLP model for text generation?
I'm looking for an example of finetuning or (training from scratch) either GPT-2 or ctrl with NetTrain to generate text. Given a title, topic, url or other control ...
9
votes
1answer
154 views
How to implement BlazeFace?
BlazeFace is a lightweight and well-performing face detector tailored for mobile GPU inference. It runs at a speed of 200-1000+ FPS on flagship devices. How to implement BlazeFace in Mathematica?
9
votes
2answers
257 views
Can I use Mathematica to generate candle scents?
I found this blog (which is highly entertaining). It has an interesting post where the author uses the GPT-2 neuro-network with candle scent training data to generate new candle scents. The AI comes ...
4
votes
0answers
87 views
Neural network for a simple game?
I'm taking a class that is based on the Mathematica coding language, and we must do a 6-week final project. I am considering coding a neural network for a simple game like 2048, pacman, snake, etc. ...
7
votes
1answer
230 views
Getting cell edges from a picture
I would like to write a program that, given a picture of an epithelium (2D cell array), for example
automatically detects the cell edges and returns the corresponding lattice. Illustratively,
...
4
votes
1answer
88 views
Adding Noise in Neural Net via ElementwiseLayer
I am trying to add noise to a neural net to train a denoising autoencoder. I don't want to pre-calculate the noisy data. Rather, I'd like the noise be applied via a function using the ElementwiseLayer....
1
vote
0answers
23 views
Training with GPU and tuning parameters with ParallelTable
We intend to train a neural network. It has a hyperparameter, such as the learning rate. We want to compare the training results of different hyperparameters. Our computer has two multi-core CPUs and ...
2
votes
0answers
55 views
Fuzzy Logic and Neural Networks
Is anyone working on building a Fuzzy Logic module that can be used for the output of Neural Networks?
Thank You
Michel
6
votes
0answers
143 views
How can I prevent errors with ReplicateLayer[]?
I am trying to build a NetGraph[] that takes input of the form {{0.6},{0.6},...,{0.8},{0.8}} and multiplies the last sequence ...
1
vote
0answers
119 views
is the code of this reinforcement learning example correct?
I am trying to learn reinforcement learning from an article on Wolfram.com, and after several days I still cannot quite get everything right. Now I wonder about the correctness of a piece of code ...
2
votes
1answer
63 views
How can I use ThreadingLayer to perform operations with layers of different dimmensions?
Is there a way that I can achieve subtraction between a scalar and a vector within NetGraph[], similar to what happens when I perform ...
1
vote
0answers
36 views
NetTrain with Quadro
I have an Nvidia Quadro P620/PCIe/SSE2 graphics card on my linux box (Ubuntu 18.04):
<<CUDALink`
CUDAQ[] (*True*)
CUDADriverVersion[] (* 390.116.0 *)
But ...
0
votes
1answer
99 views
Does lower case n have some significance in the Wolfram Language? [duplicate]
I am truly confused with this one, and maybe someone could help me to clear this up.
I am running a Table[] that runs multiple neural networks in series on OpenAI'...
3
votes
1answer
70 views
Controlling learning rate decay or cyclic learning rate in Mathematica's neural nets
I am using Mathematica's NetTrain[] function for training neural nets. There is a way to set the learning rate (LearningRate and ...
14
votes
3answers
415 views
Importing a grid of numbers from an image (sudoku like)
Can anyone see a way to import a grid of numbers from
http://www.mit.edu/~puzzle/2019/puzzle/lantern_festival.html
into Mathematica? TextRecognize doesn't seem ...
2
votes
1answer
133 views
Rounding the output Vector of a Neural Network layer to prepare it for the next layer
I am attempting to Round up the output Vector of a Neural Network layer to prepare it for the next layer
Ex. Layer output {.1.,2.2,3.3} and I need it to be a Vector of Integers for the next layer.
I ...
1
vote
0answers
84 views
How to make my custom neural layer only apply noise during training (similar to how DropoutLayer behaves)?
Utilizing some undocumented Mathematica functionality, I implemented it like this:
...
4
votes
0answers
195 views
Image-based reinforcement learning with neural networks
After seeing that "OpenAIGym" is not exactly supported on Windows and playing around with https://www.wolfram.com/language/12/neural-network-framework/train-an-...
0
votes
0answers
25 views
BatchNormalizationLayer instead of data prep
I've trained a neural net to match nx7x5 matrices to nx1 scalars; the n is a number of elements in a sequence. To get a good match I have to do some data prep; basically do a ...
6
votes
0answers
165 views
Manually converting TensorFlow models to Mathematica
Is
Mathematica (v12, my third, semi-informed, attempt at manual conversion)
...
