All Questions
682 questions
1
vote
1
answer
84
views
Training a neural network on two dataset simultaneously
I have my neural network model and want to train it at the same time on 2 datasets that share the same structure. What I have done so far does not work.
This is my code
...
- 516
1
vote
0
answers
29
views
Is this a reasonable way to transform simple MSE loss function to weighted MSE for neural networks?
The idea is tested on a dataset with a clear outlier (see picture below)
...
- 516
0
votes
0
answers
32
views
Combination of decoder and encoder for VAE interpolation
To get VAE interpolation I need decoder and encoder to be combined. Can you make a suggestion?
...
- 11
0
votes
1
answer
59
views
Scheduled Updating of LossFunction Scaling in NetTrain
I'm looking for a way to externally manipulate (hyper)parameters of a network during NetTrain training, as a function of the training progress. Specifically, I want ...
- 13
0
votes
1
answer
57
views
Use of ports on basic machine learning regression
I have this simple code to do a regression on a list of points very close to a parabole. It is purely anecdotal that I happen to know the exact underlying trend, the idea is just to have simple/nice ...
- 516
6
votes
5
answers
1k
views
0
votes
1
answer
62
views
Neural network that augments the input with constants
Suppose we have a simple neural network as follows:
...
- 487
0
votes
0
answers
97
views
SoftmaxLayer with one-hot vectors
The quantum physics problem I am looking at is a rotated qubit. The setup involves a y-rotated qubit measured in the z-basis (hence spin-up and spin-down). This scenario typically includes modeling ...
- 271
0
votes
0
answers
53
views
Using NetMeasurements for basic neural network
I have a neural network that works fine, but I would like to obtain some results on the Validation Set which provides some metrics on the performance of the training. I am obtaining some strange ...
- 271
7
votes
3
answers
253
views
How to identify and select the first name from a string containing a first name and a last name?
Is it possible to identify and select the first name in a string that is made up of a first name and a last name, e.g.:
...
- 446
1
vote
1
answer
164
views
How to do BayesianMinimization in 2D?
Studying a new experimental BayesianMinimization command in 14.0 and following an example from the documentation
...
- 29.1k
3
votes
0
answers
83
views
How to import huggingface dataset files?
The .hf file from HuggingFace's datasets library is a very popular format: https://github.com/huggingface/datasets. Is there an easy way to import these from either files on disk or their webportal?
- 3,820
0
votes
0
answers
67
views
Using CrossEntropyLoss function for basic neural network
Hi I am interested in understanding how to use the neural network functions in Mathematica. I have just started learning hence this is a basic query. The simple example I am looking at is as follows: ...
- 271
58
votes
5
answers
8k
views
Mathematica command that allows it to read my intentions
In version 2.0 of Mathematica there was a command, ReadProgrammerIntentions[ ] which allowed it rapidly compose code that could solve problems simple to complex, ...
3
votes
1
answer
161
views
0
votes
1
answer
84
views
Gradient of tensor expressions
I would like to compute gradients of tensor expressions without having to instantiate particular tensors and for arbitrary sizes.
Example 1 :
For example, let's say I want to minimize $C=\sum_j(\sum_i ...
- 101
3
votes
1
answer
89
views
0
votes
0
answers
75
views
Error following upgrade to Mathematica 14.0: Classify: TargetDevice -> GPU requires restart [duplicate]
I recently upgraded Mathematica from 13.3 to 14.0 on my workstation. When I attempted to execute a script that ran without any issues on 13.3, I suddenly get an error associated with the Classify ...
- 101
84
votes
8
answers
15k
views
What are the Wolfram Language's relative strengths for machine learning?
I see a low use of Mathematica in Kaggle competitions.
Why would one use the Wolfram Language versus R, Python, or Julia for machine learning?
Besides prettier plots and the Manipulate function, ...
- 2,403
3
votes
1
answer
77
views
StandardizedVector returned by FeatureExtract/FeatureExtraction is not the same as Standardize
The documentation would suggest that the "StandardizedVector" option of FeatureExtract and FeatureExtraction should give the same results as Standardize ("numeric data processed with ...
- 3,388
1
vote
0
answers
38
views
TensorProduct in a FunctionLayer cannot compile correctly
My network can output an array of 6x512x512 size at the end, now I want to multiply an array of length 3 for each of these 6 channels to get the color of the different channels, I am using the ...
- 83
3
votes
0
answers
136
views
What will be the next backend for machine learning?
Apache MXNet has been the choice of Mathematica Neural Network backend, but it has long since been mostly ignored by the ML community and now is finally in the attic (https://attic.apache.org/projects/...
- 2,031
12
votes
4
answers
2k
views
Text recognition - extracting table of text from image
I want to extract table from the image bellow as a list of strings. I tried the following but it does not work.
...
- 25.1k
0
votes
0
answers
36
views
Modeling with different outputs size
I have lists of input and output. I want to perform the prediction operation but my outputs are different for each input and this causes an error. Can someone help me to write the model in such a way ...
- 195
1
vote
0
answers
25
views
How to evaluate text-to-speech models in WL
There are a many solid options for text-to-speech (TTS) now in SpeechSythesize and from API providers in general. How might I use Mathematica to compare the quality ...
- 3,820
18
votes
2
answers
1k
views
Diffusion probabilistic model in deep generative modeling
Diffusion Models (http://proceedings.mlr.press/v37/sohl-dickstein15.html and https://arxiv.org/abs/2006.11239) are generative models gaining popularity in the community of deep generative modeling, ...
- 835
14
votes
1
answer
2k
views
M1 Apple silicon ML performance
Testing out a new M1 Macbook pro with Mathematica 13.0.0. I verified that I used the correct version and it is running natively on ARM.
Mathematica in general runs great, and much fast than on my ...
- 141
5
votes
1
answer
276
views
Using AI to get the geographic locations of a list of banks
I have a list of banks from Bloomberg, and am looking to locate them on the world map,
...
- 2,223
1
vote
0
answers
96
views
Why does ridiculous error information appear in Classify with Method GradientBoostedTrees?
The data could be classified correctly by method like LogisticRegression, NearestNeigbors.
The dataset is quiet simple, ...
- 8,649
2
votes
0
answers
54
views
Publishing image classifier as a website or to the cloud
I have made an image classifier by importing images and training my neural network, and I would like to share this classifier with my colleagues that don't have access to Mathematica.
However, if I ...
6
votes
1
answer
542
views
How do I generate SMILES using machine learning in Mathematica?
I am wondering if anyone has experience with generating molecules using machine learning in Mathematica. There are a bunch of papers these days with such applications/developments. For example, https:/...
- 739
91
votes
1
answer
14k
views
Crack CAPTCHA using deep learning
Deep convolutional neural networks are very good at computer vision related tasks. Optical Character Recognition (OCR) is one important branch of computer vision. In fact, the convolution neural ...
- 27.8k
1
vote
1
answer
51
views
FeatureSpacePlot output differs in documentation vs. a notebook
I was exploring this first example of FeatureSpacePlot and it shows some strange behavior. On the left is the documentation window and to the right is the notebook. ...
- 59.6k
0
votes
0
answers
26
views
Neural Network erros in NetEncoders?
I'm making a fairly complex neural network, and I've listed all the ports, the network flow, and the layers well. However, when I tried to run the NetGraph, it gave me the error message
...
0
votes
0
answers
68
views
How to extract images from Information?
class = Classify[
ExampleData[{"MachineLearning", "UCILetter"}, "Data"]]
Information@class
I have queried the Information function, ...
- 1,630
17
votes
1
answer
616
views
Stable diffusion models in WL?
Are there any diffusion-based (text + image) -> image generation models available in Mathematica, Wolfram Repositories, or from third parties?
In particular, I'm ...
- 31.8k
3
votes
1
answer
318
views
Prediction Algorithm [closed]
Trying to identify what algorythm I need to use to predict a battery full time (100%), based on historic data (charge rate is variable).
I suspect it's some form of linear regression of a sort, ...
- 31
3
votes
1
answer
118
views
Evaluating a questionnaire from image
I want to evaluate a questionnaire made at school. The survey is made on paper and I will scan it. So I have an image with all the questions (44 questions). What I am looking for is which number was ...
- 1,590
48
votes
5
answers
4k
views
Generative Adversarial Network
Generative adversarial networks (GAN) is regarded as one of "the most interesting idea in the last ten years in machine learning" by Yann LeCun. It can be used to generate photo-realistic images that ...
- 27.8k
5
votes
1
answer
198
views
Does Mathematica have a LogSumExp function that's numerically stable?
I searched "Mathematica LogSumExp", and literally nothing is coming up. It's a relatively important function in machine learning and for numerically stable computation. Will I just have to ...
- 191
0
votes
1
answer
61
views
Creating an If function for the probabilities of the classes of images after using NetTrain
So I have 3 Classes the model can output after veiwing an image, at the moment it has a 70% accuracy, however this is not an issue as the plan is unless it's 85%+ sure on the Probability of the class ...
11
votes
1
answer
490
views
Can LearnDistribution learn conditionality?
I would like to apply LearnDistribution to multivariate conditional distributions. Here's a simple artificial example:
...
- 31.8k
9
votes
2
answers
803
views
A strategy to create good initial guess for Lorentzian model fit
I was preparing an answer for a question that got deleted, so I will post this anyhow.
Given a set of data with a hump
...
- 37.4k
10
votes
1
answer
380
views
Labeling images for TrainImageContentDetector
I want to use TrainImageContentDetector on a directory of ~2k unlabeled images. I'm willing to put in the effort to label them, but I was never good at using "...
- 3,820
0
votes
0
answers
65
views
How to use NetChain to obtain a fit that is at least accurate on the training data
How do I use NetChain to obtain a fit that is at least accurate on the training data?
I am trying to get a hopefully predictive fit for some data (see my question ...
- 4,781
2
votes
1
answer
175
views
CrossValidate for image classification with NetTrain
I am trying to set up transfer learning using instructions from this repository page on EfficientNet. I am able to perform a stand-alone training and testing using the instructions on that page. The ...
- 739
73
votes
6
answers
5k
views
Q&A Mathematica v.11 Neural Networks: A comprehensive look at Layers, Net Functions, and pioneering into this [[experimental]] code
Motivation
As Mathematica v.11 was released earlier this month with a host of new [[experimental]] functions and a limited number of examples on curated data that ...
- 5,442
6
votes
0
answers
287
views
Any competitive speech-to-text models?
SpeechRecognize is not very useful compared to most voice transcription engines nowadays, simply too fragile and prone to errors.
I use Otter.ai currently, is there anything competitive to that (or ...
- 3,820
1
vote
1
answer
153
views
Generating molecules using a GAN in Mathematica
I have been unable to find any tutorials or examples of creating molecules using a Generative Adversarial Network (GAN) in the Wolfram language. It appears that the Selfies paclet was created to ...
- 739
48
votes
2
answers
9k
views
Object detection and localization using neural network
Four important computer vision tasks are classification, localization, object detection and instance segmentation (image taken from cs224d course):
These four tasks are all built on top of the deep ...
- 27.8k