Tell the library (similar to the "MICE" package in R) for Python which implements methods for recovering gaps in data.

Closed due to the fact that it is necessary to reformulate the question so that you can give an objectively correct answer to the party Nicolas Chabanovsky ♦ 14 sep '16 at 5:32 .

The question gives rise to endless debates and discussions based not on knowledge, but on opinions. To get an answer, rephrase your question so that it can be given an unambiguously correct answer, or delete the question altogether. If the question can be reformulated according to the rules set out in the certificate , edit it .

  • sklearn ? - MaxU
  • In the sklearn implementation of these methods for the restoration of gaps there. Are there any other options? - Taras
  • What do you mean by “restoring passes” - prediction, interpolation, something else? - MaxU
  • I mean prediction. Now I replace the missing values ​​in the data with 0 (which is the easiest way) and I want to improve the results by replacing the gaps with a more complex method. I found a description of two ways to do this: regression models and the EM algorithm. But I did not find their implementation in Python. - Taras

1 answer 1

I'm not sure about the EM-algorithm, but the regression methods are presented in sklearn quite well :

SGD Regressor

Lasso

ElasticNet

SVR kernel = 'rbf'

Ensemble regressors

Ridge regression

SVR kernel = 'linear'

  • This is not what I need. I need a function that sends a frame with missing data to the input, and we get the same frame to the output but with the gaps filled in - Taras
  • "skips" can be in many (all) columns? - MaxU
  • Yes, passes can be in many columns - Taras
  • Look at the Imputer , the truth is rather filling with average or median values, and not prediction - MaxU
  • I already looked at Imputer. there are the most simple functions, but this is not what I'm looking for - Taras