quester
  • Member for 3 years, 11 months
  • Last seen more than a month ago
Should I use regex or machine learning?
2 votes

If in case 1 there would be one template for organizing information on CVs then you can go for regex, but in order to have really helpful tool for real world CV you have to train ML solution. Just ...

View answer
Suggestions on how to explain 'models' & 'algorithms'
1 votes

maybe not definitions but more intuition: algorithm is a recipe to do something (train model, find max-cut etc.) model is a mathematical description how objects interact/bahave in given environment ...

View answer
Is it possible to identify what data a neural net was trained on?
1 votes

If you can see the loss for a specific dataset, then the most probable set would be the one with the lowest loss.

View answer
Aggregate several attributes into one using PCA
1 votes

PCA will change your data and you will not be able to interpret it in sane sense, you can just slice and dice the data and do many things by hand, PCA would be usefull if you would want to find "...

View answer
Is it possible to have a default class in multi class classification?
1 votes

just use predict_proba and in the end if p is lower than some threshold (for 10 classes all p will be ~.1 and best could have 0.2 so if it's to low for you) you will just change predition from "class1"...

View answer
How does Neural Network denoise an image?
1 votes

in general CNN can generate images, so we can feed them noised images and images without noise at output and it can learn how to figure out pixel color based on his neighbours, actually you can't ...

View answer
Classification vs Regression Model what should I choose?
1 votes

if you can apply dummy operation like if ... else ... to get categories then choose regression since you have to predict revenue, also it will "tell" your model that some cutomers will give very big ...

View answer
How to deal with a feature that has lot of categorical values?
1 votes

one idea is to craft some features of this feature - when you classify cars you don't have data like strings "Ferrari 991 year 2014 red" "BMW z4 year 1999 2.0L blue" but you would like to have columns ...

View answer
good algorithm for outliers detection
0 votes

what do you mean "outliers"? if you mean new behavior based on numbers then just go row by row, see if new observation is in some safe bounds based on historical data and update these bounds based on ...

View answer
Why do we need convolutions over volume in convolutional neural networks for image recognition?
0 votes

in medical images often colors are very vivid and in channels looks very different take a look at these contests data https://www.kaggle.com/c/human-protein-atlas-image-classification https://www....

View answer
How to form and minimise custom features for classification in supervised learning
0 votes

https://en.wikipedia.org/wiki/Support-vector_machine polynomial kernel of degree 2, with C coefficient >.1 but you would have to figure it out based on your data I can provide more info on request

View answer
Cross-fold validation done on whole dataset or training set?
0 votes

do bagging and then evaluate model, also you should split your data into train and test 60/40 and on train do folds or split it into train and validation https://en.wikipedia.org/wiki/Training,...

View answer
Data splitting for a binary classification model
0 votes

if this is how your data looks like it would be a good idea to split users into training examples and test examples, train examples would contain users with info about all related apps, and in test ...

View answer
ML model deployment architecture?
0 votes

Just treat the model as you would treat an image or other resource in your project. Would you test images at staging too?

View answer
How to implement feature selection for categorical variables (especially with many categories)?
0 votes

try to cluster them - like when you will have several small cities nearby maybe it would be better to treat them as one city like Seattle and Bellevue could be combined into one category Seattle

View answer
Using LSTM for binary text Classification, getting almost same accuracy at each epoch
0 votes

because it shows score after each batch and in the end of epoch score is 0.74, don't look on scores after each batch since data can be shuffeled like this: first batch contains examples that your NN ...

View answer