Questions tagged [machine-learning]

Methods and principles of building "computer systems that automatically improve with experience."

Filter by
Sorted by
Tagged with
1
vote
0answers
5 views

How does a Faster R-CNN gets trained and how does the Region Proposal Network fits into that?

so a Faster R-CNN consists of 3 parts: -the Base CNN -the Region Proposal Network with RoIpooling -and the classification CNN at the end So I asked myself if all these three Networks are trained ...
0
votes
1answer
8 views

What are the factors to consider when setting the depth of a decision tree?

In scikit learn, one of the parameters to set when instantiating a decision tree is the maximum depth. What are the factors to consider when setting the depth of a decision tree? Does larger depth ...
0
votes
1answer
6 views

Why are bigger embedding vectors not necessarily better?

I'm wondering why increasing the dimension of a word dimension vector in NLP doesn't necessarily lead to a better result. For instance, on examples I run, I see sometimes that using a pre-trained 100d ...
0
votes
1answer
14 views

How will a decision tree cope if it cannot find a suitable feature to choose as root node from which to split further?

When I look at decision trees, they start with a root node choosing the most suitable feature from which to split further. What if the decision tree is unable to find the most suitable feature from ...
0
votes
0answers
7 views

Quadratic programming and Lagrange multiplier in SVM

I am a little confused because for some simple functions and constraints, using the Lagrange multiplier will be able to solve for the variables. However, in the SVM Lagrange expression, I learned that ...
0
votes
0answers
8 views

About SVM cost function and Lagrange

I just watched the MIT video on the intuition and mathematics behind SVMs, and overall, I learned that the objective is to minimize the margin or the distance between the support vectors with some ...
0
votes
0answers
9 views

SVD++ in matrix form

I have been recently reading the following presentation and papers preceding it: http://staff.ustc.edu.cn/~ynyang/group-meeting/2014/matrix-factorization/Matrix-factorization-recommendation-system-...
0
votes
0answers
7 views

Ideas to derive new column or eliminate a column for Logistic Regression problem related to binary classification

I am working on a Logistic Regression problem to identify the leads that result into a student conversion for candidature. I have 2 columns in the dataset 'Last Activity' and 'Last Notable Activity' ...
0
votes
0answers
7 views

Using Unsupervised / Supervised Learning for RGB manipulation

I need to teach a model(or any substitute) for automatically adjusting the concentration of R, G, and B values until they make up white colour. i.e, tweak the RGB values (beginning from some arbitrary ...
0
votes
0answers
14 views

How to label images for a task similar to AWS Textract?

I am trying to build a model to detect tables from images of machine plates(PFA) and create key/value pairs of the information present in the table. But I am not very sure that what is the best way to ...
0
votes
0answers
7 views

Validation of Images generated using DCGAN

I have trained a Deep Convolutional Generative Adversarial Network(DCGAN) model and generated some images. Now, I need to validate these images if generated images ...
1
vote
1answer
24 views

How to use a loss function that is not differentiable?

I am trying to find a codebook at the output of a fully connected neural network which chooses points such that the minimum distance (Euclidean norm) between the so produced codebook is maximized. The ...
0
votes
0answers
7 views

how to do topic modeling on very huge data?

When i come to know that gensim is useful library for topic modeling, I tried it on my huge amount of document. It works well only if the dictionary size is to be fix. In my case i have each and every ...
0
votes
0answers
9 views

Dataset throwing KeyError while looping through a list of variables [migrated]

I have a list of variables with values encoded in a way which throws Pandas off. For example: I have a column named "Alley" and it has a list of values, one of which is ...
0
votes
1answer
16 views

Can we specify the number of data generated(minority class) using SMOTE?

I am trying to improve classification of imbalanced dataset creditcard fraud using SMOTE imbalanced_learn. But, in this it generates the data to 50%, can we give a specific number for the data to be ...
0
votes
1answer
31 views

Is my model overfitting? The validation loss keeps on fluctuating

I have trained a 4 layer neural network ...
0
votes
0answers
7 views

Suggestion to use transfer learning in detection of Environmental norms violator [on hold]

I am working on one research, where the goal is to predict if a particular company will violate environmental norms in future from their websites screenshots. So, this goes to be binary ...
-1
votes
0answers
7 views

What is the best method for classification of time series data Using RNN GRU

Data which has a few input signals which are numeric and output signal which are Categorical, like, Rainy , Sunny, Snow . So its a multi category classification. If I am using RNN GRU , what would ...
2
votes
0answers
29 views

Structuring a LSTM Layer

I'm trying to improve an NER Bert sequence tagger using LSTM layers in TensorFlow. I'm a bit unclear on the interface and how a LSTM layer should be set up. Currently, I'm taking in 3-5 sentences and ...
2
votes
2answers
39 views

Procedure for selecting optimal number of features with Python's Scikit-Learn

I have a dataset with 130 features (1000 rows) . I want to select the best features for my classifier. I started with RFE but Its taking too long, i done this: <...
0
votes
1answer
15 views

Which algorithm to use to classify plots based on graphical features?

I have plots which I need to classify based on some features. For example, I need to differentiate between the following plots having smooth features or 'valleys' at certain x values. Which machine ...
1
vote
0answers
27 views

Semantic Segmentation Networks incorrectly guessing single pixel values for all masks

So, I've been running a test to see how well a number of networks can perform road segmentation on a particular customer's dataset. I am testing UNET, RDRCNN, and Tiramisu against each other. UNET ...
1
vote
4answers
54 views

Neural net only works for small datasets

I have a neural network that attempts to solve some regression problem. The network works fine when the dataset has a small number of training examples, lets say 20. It overfits of course, but the ...
2
votes
2answers
23 views

How to use machine learning to create combine of opposite images side by side

Inspired by: Two Worlds Pictures I just want to create a Machine Learning Model that can automatically combine the opposite images into 1 image. I am thinking about 2 possible solutions: Pose ...
2
votes
1answer
24 views

SGD, calculating it by hand

While I find a lot of material of SGD (Stochastic gradient descent), I am struggling to find one concrete example with numbers e.g. calculating it by hand for let's say, one iteration would help me a ...
0
votes
2answers
31 views

How to convert trained data (feature extracted) into a prediction model? [on hold]

Background: I am analyzing and labeling some log data. (parsed already, sample data below) I have extracted the major features of data. For examples, the classification results ("1 - normal" or "0 - ...
2
votes
2answers
65 views

What regression model can handle tiny amounts of data?

I'm trying to use machine learning to predict properties of a material during a crash test, but each data point requires physically crashing an expensive toy car, so I can only gather a few hundred ...
0
votes
0answers
9 views

I have a question about the structure of Faster R-CNNs

So hi, I need some help to understand the whole structure of the Faster R-CNNs. Further i will list the steps of the structur with questions between them: Image -> CNN -> (is the output ONE or ...
0
votes
0answers
2 views

TPOT freezes while optimizing in Ubuntu 18.04 [migrated]

I have started working with tpot with dask. But the optimization always freezes at 0%. I have tried almost all the solutions suggested. But still no luck. Here is my code - ...
0
votes
0answers
15 views

LSTM Cross Validation with Rolling Window

I'm stuck in coding a cross-validation technique for my LSTM RNN. My supervisor advised me to go with a rolling window cross-validation. My goal is to compare the RMSE for same day predictions. I am ...
1
vote
2answers
18 views

Equation related to Smoothness

If you have a differentiable function $f:\mathbb{R}^d\rightarrow\mathbb{R}$ that is $\beta$-smooth (for all $v$ and all $w$, you have $\|\nabla f(v)-\nabla f(w)\| \leq \beta \|v-w\|$), how can you ...
1
vote
0answers
25 views

Mobile App Recommendation: How to get the rate of a specific user submit for a specific application

I have a mobile app recommendation project, so I need data set which has user-app matrix-rate. Actually, I want to know what rate does a specific user submit for a specific application. in other words,...
0
votes
0answers
11 views

Fully Convolutional Variational Autoencoder

I would like to make a neural network which uses black and white images as input and outputs a colored version of it. The important thing in that process is that the size of the images must stay the ...
0
votes
1answer
15 views

Which is the correct method for outlier analysis on a dataset for modelling?

I'm trying to build a regression model but my data-set have many outliers points which I need to analyze and then remove them. There are two ways to do it, 1) First do all the analysis on every ...
0
votes
1answer
22 views

What should be the criteria to select features using correlation factors between features? [duplicate]

For the titanic dataset, I have done some feature engineering (one-hot encoded the features) and now I have developed a heatmap to view the correlation between different features. I'm not able to ...
0
votes
1answer
17 views

Contextual Spell Correction

I want to create a spell checker that corrects the spelling mistakes contextually. For example, Erroneous sentence: I want to apply for credit cart Corrected sentence: I want to apply for credit ...
0
votes
1answer
9 views

Batch-mode active learning: How to select the batch that will bring the largest model improvement

I am facing a problem where I want to use active learning to improve my classifier. Basically, I can choose data from one (and only one) data set among a set of candidate data sets. The question is ...
0
votes
0answers
8 views

Python and CPLEX [on hold]

I am working with a project is about Software Defined networking. We consider for optimization problem. I use python programming language with mathematical optimization solver CPLEX. We use k mean ...
1
vote
1answer
13 views

Unable to learn weights of a Word2Vec model

I was going to implement a word embedding model - namely Word2Vec - by following this TensorFlow tutorial and adapting the code a little bit. Unfortunately, though, my model won't learn anything. I've ...
0
votes
0answers
19 views

What kind of NN architecture should I use to my problem?

I'am new to Data Science, and I'am trying to figure out how to solve my learning problem: My data set is CIFAR-100. Each image in the set got a vector of properties, a vector on $\mathbb{R}^n$. My ...
0
votes
0answers
15 views

KDTree classified as AI /ML?

We are currently using a KDTree to find similar documents based on keywords identified via TF-IDF. After the tree is built, we are querying the tree to find nearest points. Would this fall in the ...
2
votes
1answer
25 views

How to find the probabilities of certain events occurring in a defined sequence?

Good day, everyone! I have a problem where I have to program something like this: I have some arbitrary number of events, let's call then Event A, B, C, D, E, F, ... and so on. Now they occur in ...
0
votes
0answers
7 views

Backtracking Line search for Multiclass classification gradient descent

For my case i am dealing with multiclass problem and there are total 28 direction component for each class and there are total 5 classes, for given equation above, f(w+nd) and f(w) gives scaler values ...
0
votes
1answer
40 views
+50

use ANN to predict several properties

Let’s say I have tens of thousands of datasets (rows) consisting each of 300 columns of integer, double and strings (no image, no audio). Five of these target columns represent interesting resulting ...
0
votes
4answers
30 views

How to know if my Decision tree model is good or bad?

I have built a decision tree model. I am not sure if it is good or bad as i am new comer in ML. Kindly help to evaluate my model" Here is my code which i used, ...
0
votes
0answers
8 views

In learning latent factors in matrix factorization, are weights learned based on similarity between e.g. users?

I implemented a recommendation system using user-user interaction data, learning missing ratings through alternating least squares and matrix factorization, which as I understand it, adjusts and ...
2
votes
1answer
20 views

How to normalize a position in 3d game?

I'm pretty new to anything and everything related to this kind of stuff, I was wondering how would I normalize the coordinates of a entity in my game for a nerual network? Would it just be the same ...
1
vote
1answer
21 views

very important features, but rarely input

The features are of capital gain and capital loss, but very small amounts of people have one or the other. As far as I can see it may not signify any great difference, though I'm not sure how I would ...
0
votes
0answers
3 views

How to design a distributed system to deal with X amount of data given Y resources? [closed]

How to design a distributed system to deal with X amount of data given Y resources?