pplonski
  • Member for 6 years, 1 month
  • Last seen more than a month ago
Do Random Forest overfit?
12 votes

The Random Forest does overfit. The Random Forest does not increase generalization error when more trees are added to the model. The generalization variance is going to zero with more trees used. I'...

View answer
How do subsequent convolution layers work?
2 votes

Check this lecture and this visualization Usually it is used type 2.1 convolution. In the input you have $NxMx1$ image, then after first convolution you will obtain $N_1xM_1xk_1$, so your image after ...

View answer
How to update weights in a neural network using gradient descent with mini-batches?
2 votes

When you train with mini-batches then you have the second option, network is updated after each mini-batch, and epoch ends after presenting all samples. Please see these responses

View answer
Can I fine tune the xgboost model instead of re-training it?
Accepted answer
2 votes

I see that in the current version of python wrapper of xgboost you can specify file name or existing xgboost model (class Booster) in train function.

View answer
What are some nice algorithms/techniques for optimizing and predicting Click Through Rates (CTR)?
2 votes

You can see nice algorithms in kaggle competitions about CTR: https://www.kaggle.com/c/avito-context-ad-clicks https://www.kaggle.com/c/avazu-ctr-prediction Just go to forum of each competition and ...

View answer
RandomForest - Reasons for memory usage / consumption?
0 votes

The memory usage of the Random Forest depends on the size of a single tree and number of trees. To control the memory size of RF you can: limit the number of trees (the dependency is almost linear), ...

View answer
overfit a Random Forest
0 votes

I was doing very similar exercise. I've generated the synthetic dataset: y = 10 * x + noise and fitted one Random Forest model with full trees and one with pruned: # ranadom forest with full trees ...

View answer
Can regression trees predict continuously?
0 votes

In classic regression trees you have a one value in the leaf, but in the leaf you can have a linear regression model, check this ticket. You can also use ensemble of trees (Random Forest or Gradient ...

View answer
Can PyLearn do everything that Theano can?
0 votes

PyLearn2 is built on top of Theano. If you have Theano snippets you can run them with PyLearn2 snippets.

View answer
Random Forest Regression. How to represent really long list of categories for processing
0 votes

I think that your first and last approach are the same - as a result you have sparse feature vector. In your last approach you just don't list features with value 0. It is hard to say if you should ...

View answer
What is the appropriate evaluation metric for RandomForest with probability in R?
Accepted answer
0 votes

You should select some threshold, let's say 0.5 and treat customers with probability below threshold as not buy and above as buy. Based on this you can compute accuracy of your model. You can also ...

View answer
Training and testing AdaBoost for low probability classification
0 votes

I think that it depends on your data set. There are many ways to handle unbalanced data sets, just search, for example https://www.quora.com/In-classification-how-do-you-handle-an-unbalanced-training-...

View answer