training (or estimation) of statistical models or algorithms.

learn more… | top users | synonyms

0
votes
0answers
23 views

Inverse progression for training & validation data during training with H2O

I'm training on a dataset with 3600 columns. 100948 training rows & 25238 validation rows. These are the R commands I'm using: ...
1
vote
0answers
42 views

How to train radial basis function for function approximation?

There is an Autoregressive model of order 1 (AR(1)) that is excited by a non-linear signal as the input: $$x_t = \rho x_{t-1} + u_t \tag{1}$$ The time series $u_t$ is generated from a Mackey-Glass ...
1
vote
1answer
31 views

Is this training dataset enough for training and testing classification model?

My training dataset contains just 2 classes with 40 features. In case 1, class 1 has 35 samples and class 2 has 700 samples. In case 2, class 1 has 65 samples and class 2 has the same value as ...
0
votes
0answers
32 views

Training set selection

I have the following question for a project I'm working on. I am trying to find the best strategy to select the best training set in a dataset. I have a dataset with a few billions rows. I am trying ...
0
votes
1answer
69 views

Creating a test set with imbalanced data

I am working on a binary random forest using R. mu data set consists of 300 cases classes 1 and 2100 cases class 0. I am planning to evaluate my model using the model prediction and the AUC and for ...
1
vote
0answers
33 views

too many ties in knn? how to solve this problem

I use the knn model to train my data and then eliminate accuracy via cross-validation, but when i use the following code, I get the error: Error in knn3Train(train = c(1680, 300, 480, 240, ...
0
votes
0answers
26 views

How to detect contradictory examples in training data

I want to implement an online learning classification algorithm. Therefore I want to detect contradictory examples in my training data once the user wants to add a new example to the training data. ...
0
votes
0answers
51 views

TfidfVectorizer: should it be used on train only or train+test

When training a model it is possible to train the Tfidf on the corpus of only the training set or also on the test set. It seems not to make sense to include the test corpus when training the model, ...
0
votes
0answers
81 views

Random Forest online/incremental learning in R

Is there a Random Forest implementation available in R, that supports online learning? My alternative approach was to use the popular randomForest package and combine Random Forests (the existing one ...
1
vote
0answers
45 views

Test MAPE < Train MAPE using auto.arima()

I am trying to build a forecasting model for the passenger vehicles registrations in a given country, and I wanted to use $auto.arima$ function from the $forecast$ package to estimate a simple ARIMA ...
1
vote
0answers
23 views

Should the number of normal samples always be more than that of anomalous samples in training set for anomaly detection?

I am trying to train an anomaly detection algorithm (one-class svm) on a data set with a few hundred positive samples and several thousands negative examples. Is it mandatory that I train the model ...
0
votes
0answers
14 views

Which phrase is correct? “Out of sample” or “Hold out sample”?

Which one is correct phrase for data that model didn't see? For example i trained my model based on data-set from 200 to 2002. Now i want test my model using 2003 data-set. Which phrase is correct in ...
0
votes
1answer
95 views

Extracting Standard Errors Caret Model

I have tuned a glm net model with caret using the train function. I am trying to extract the coefficients and standard errors of those coefficients for the best tuned model. Following this CV post I ...
0
votes
0answers
104 views

In machine learning, may I train correctly a neural network with input real data and output validation Boolean data?

I have a matrix made of ~ 100 rows and 12 columns. Each entry contains a real value. The first 6 columns refer to a particular concept (firstClass), the following 6 to another one (secondClass), and ...
1
vote
0answers
16 views

Learning over Multinomial data

I have a training data with 68 features... Each of which is a different multinomial distribution. Eg. Feature 1 can take 1 of 4 values while feature 2 can take one of 10 values. Which classifier or ...
0
votes
1answer
31 views

Cross-Validation in binary classification using only 10 positive samples (SVM)

I have a binary classification problem for which only $10$ positive samples are available for training. Negatives are in general in abundance, but I choose to use solely $70$ ($7$ negatives per one ...
0
votes
1answer
62 views

How to divide dataset into training and test set in Recommender Systems?

I am working on two simple recommender systems - Collaborative (item-item a user-user) and Content Based. I would like to evaluate prediction accuracy of these systems. I am used to divide dataset ...
4
votes
1answer
57 views

How to train a model when instead of a target we have a range where it is?

Often in machine learning we have a situation when target is numeric (real or integer). Each target comes with an associated input vector. The goal is to learn the mapping from the input vectors to ...
2
votes
0answers
1k views

Different results from randomForest via caret and the basic randomForest package

I am a bit confused: How can the results of a trained Model via caret differ from the model in the original package? I read Issue on prediction with FinalModel of RandomForest in R using the CARET ...
1
vote
0answers
51 views

Is there any relationship between Train error and Test error in linear regression?

I am doing regularization on linear regression and am observing something arguable: By increasing the $\lambda$ (shrinkage value), I observe that there is a point in which the training and testing ...
2
votes
2answers
55 views

Should I use epochs > 1 when training data is unlimited?

If I have virtually endless training data (it's synthesized) is there still purpose in having epochs? I.e. training on the same samples multiple times?
0
votes
0answers
26 views

What is the best practices as to cropping the positive examples for training hog classifier?

I've been reading up on the literature on HOG training, but I can't figure out what the best practices are for cropping the positive examples. Question: Do you crop tight or with margins on the ...
2
votes
1answer
141 views

Test Error less than cross-validation error-implications?

If the test-set RMSE error of a model is less than cross-validated RMSE error, how can I interpret this? Is this abnormal? Does it imply a mistake in the methodology?
0
votes
1answer
165 views

training and testing an artificial neural network method

Is it alright to apply different epoch numbers to train/test a ANN method ( i.e. set number of iterations/epoch for training mode, and then set another different number of epoch for testing mode) ?
0
votes
0answers
69 views

Can you take a DNN that was trained without regularization, and continue training it with regularization?

If I've trained a DNN with out any regularization methods (e.g. weight decay, dropout etc.) and reached a good training error, can I somehow take that learned net and fine tune it with regularization? ...
3
votes
1answer
190 views

proportion between test and train set in regression and classification

Many claim different proportion between test and train set in regression and classification telling this should be test:train : 0.25:0.75 or 0:33:0:66 (the most popular with what I met). But what with ...
0
votes
0answers
17 views

Generating Labeled Training data from 2 data sources for Predictive Classifier

I am trying to build a predictive risk model classifier for an product (classifying good or bad). I am in the process of creating a training dataset. Here are the challenges I am facing. I have 2 ...
0
votes
1answer
218 views

How we can add new data in training time of neural network without stopping it in MATLAB?

I have a binary classification problem. Now I'm using patternnet in MATLAB R2014b to design a neural network for this problem. ...
1
vote
0answers
44 views

Adaptive Learning Rate Convolutional RBM?

I was wondering if anyone was aware of some work done for Adaptative learning rate for Convolutional RBM training ? KyungHyun Cho published an algorithm for RBM (Enhanced Gradient and Adaptive ...
0
votes
0answers
104 views

Training Neural Network with Highly Correlated Inputs

I am trying to train a basic Neural Network to predict Football final scores based on: i) Time in the match ii) Current Score iii) Parameters representing strength of home and away team. In order ...
0
votes
0answers
57 views

Honglak Lee MNIST CDBN filter shapes?

This question is based on Honglak Lee's paper "Convolutional Deep Belief Networks for Scalable Unsupervised Learning of Hierarchical Representations". In chapter "4.3 Handwritten digit ...
0
votes
0answers
21 views

Importance of class frequency in classification

Suppose we are classifying instances into n classes that, in practice, occur at frequencies p1, p2, ... pn, (for example classifying news-articles as one of n different topics). For the purposes of ...
0
votes
1answer
58 views

Once you have a set of training data, how do you go about training a model?

I've been trying to teach myself machine learning using some books, kaggle and other resources so I'm still very new at this. I'm trying to do a competition on kaggle and I've obtained a dataset. ...
0
votes
0answers
93 views

How to choose negative training sample for Classification problem

Choosing positives sample is a relative straightforward task, but I'm having some problem on determine what should I use for the negative example. I'm working on a SVM binary classificator, trying to ...
1
vote
0answers
215 views

Pre-training deep neural networks by supervised learning

When pre-training deep neural networks layer by layer, is it normal to pre-train the layers -which haven't been pre-trained by unsupervised training- by using supervised training before we train the ...
0
votes
0answers
54 views

semisupervised classification training on all or part of unlabeled data

I have 3 sets of data. A positively labeled dataset. An unlabeled dataset that has for sure positive (around 75%) and negative data. An unlabeled dataset that has for sure positive data and maybe ...
1
vote
0answers
357 views

Tips for training dropout neural network

I use NN for my mini project research, and I found out the newest trick for feed forward NN is using dropout for regularization instead of L1/L2 norm and rectified linear unit as an activation ...
0
votes
1answer
130 views

How to compute the weights gradients for Convolutional RBM with CD?

I've successfully implemented a RBM and trained it with CD-k, but I now have some issues doing the same with Convolutional RBM. The visible and hidden biases gradients are easy to compute: v1 - v2 ...
0
votes
1answer
101 views

how to handle (many) false positives in training dataset for logistic regression classifier

I want to train a logistic regression dataset. I have a quite big training data set ( >100 000) and have around 10 features I can train on. Half of my training data is negative training data and I ...
2
votes
1answer
39 views

Latent semantic classification

How can I create a training data set for document classification using LSA? I have created a term-to-document matrix and have class labels also. I don't know whether to add these class labels in a ...
0
votes
1answer
601 views

How to speed up the training in neural network when mini-batch training is used?

Can anyone give me some ideas on possible techniques to speed up the training process of multilayer artificial neural network if the training involves mini-batch? So far, I understand that stochastic ...
1
vote
0answers
354 views

HMM library, different length sequences training

I'm using the Kevin Murphy's HMM library in MATLAB(http://www.cs.ubc.ca/~murphyk/Software/HMM/hmm.html) There is a section called 'How to use the toolbox'. There is this example for GMM ouputs: ...
2
votes
2answers
117 views

Split train//validation/test sets by time, is it correct?

Here's the scenario, slightly altered to a common one. Credit card fraud, payments for the last 12 months (a rolling window). Train with the data from the first 10 months, validate with data from the ...
2
votes
2answers
153 views

Building training set from unlabeled data

I want to build classifier using naive bayes. I know that naive bayes is supervised learning that requires labeled data for training set. However, I only have data with no label on it. Is there any ...
2
votes
3answers
962 views

Neural network packages which allow shared weights and parallel training

I'm curious if there are any neural network packages out there that easily allow one to construct feed forward neural networks with shared weights, but also allow for the training to be done in ...
3
votes
1answer
257 views

Imputation before or after splitting into train and test?

I have a data set with N ~ 5000 and about 1/2 missing on at least one important variable. The main analytic method will be Cox proportional hazards. I plan to use multiple imputation. I will also be ...
12
votes
3answers
601 views

Training, testing, validating in a survival analysis problem

I've been browsing various threads here, but I don't think my exact question is answered. I have a dataset of ~50,000 students and their time to dropout. I am going to be performing proportional ...
1
vote
0answers
370 views

k-means + linear regression: How to split the data for validation

I want to cluster my data first using k-means and then determine a regression model for each cluster. Then I want to evaluate the performance of this approach using split validation. I can think of ...
5
votes
1answer
1k views

How to perform parameters tuning for machine learning?

I have a very basic question regarding parameter tuning using grid search. Typically some machine learning methods have parameters that need to be tuned using grid search. For example, in the ...
0
votes
1answer
69 views

Improvement on duplicating instances

I have a task of Relationship extraction. There are some set of predefined relations in the corpus. I need to train classifier to recognize the type of relation or the lack of relation between every ...