Styjun

Is there a set of positive integers of density 1 which contains no infinite arithmetic progression?

How to trick the reader into thinking they're following a redshirt instead of the protagonist?

Is it expected that a reader will skip parts of what you write?

Longest bridge/tunnel that can be cycled over/through?

How to hide rifle during medieval town entrance inspection?

Can the removal of a duty-free sales trolley result in a measurable reduction in emissions?

Explain the ending of Black Mirror's "Smithereens"

Bb13b9 confusion

How can I make 12 tone and atonal melodies sound interesting?

Fermat's statement about the ancients: How serious was he?

Why did Intel abandon unified CPU cache?

60s or 70s novel about Empire of Man making 1st contact with 1st discovered alien race

How can I end combat quickly when the outcome is inevitable?

Why am I Seeing A Weird "Notch" on the Data Line For Some Logical 1s?

Live action TV show where High school Kids go into the virtual world and have to clear levels

Printing Pascal’s triangle for n number of rows in Python

Which languages would be most useful in Europe at the end of the 19th century?

Is using 'echo' to display attacker-controlled data on the terminal dangerous?

How to “listen” to existing circuit

A map of non-pathological topology?

Next date with distinct digits

What aircraft was used as Air Force One for the flight between Southampton and Shannon?

Who won a Game of Bar Dice?

How can I remove material from this wood beam?

exchange of measurement errors in 2d array

Is there a NumPy function to return the first index of something in an array?How to print the full NumPy array, without truncation?error: Unable to find vcvarsall.batNumpy array dimensionsPeak detection in a 2D arrayHow to access the ith column of a NumPy multidimensional array?Dump a NumPy array into a csv fileMeasure time elapsed in Python?Convert pandas dataframe to NumPy arrayfatal error: Python.h: No such file or directory

.everyoneloves__top-leaderboard:empty,.everyoneloves__mid-leaderboard:empty,.everyoneloves__bot-mid-leaderboard:empty height:90px;width:728px;box-sizing:border-box;

0

I have very large evaluation data of strain readings which I import as 2d arrays into python. Short Example:

X = np.array([[210, 211, 209, 10000, 215, -5000],
 [220, 221, 219, 10000, 225, -10000],
 1400, 225, 230, 10000, 235, -98]])

Some measured values can be interpreted directly as measurement errors (e.g.:10000,-5000). I define them with an interval (e.g.: <-10, >1000). Then the values should be removed from the array and replaced by suitable values. The measurement errors should either be interpolated or replaced by other methods.

I already had a few attempts with the Panda's function asfreq() and imputation class from scikit-learn. With asfreq() I had the problem that it only fills NaN values. With imputer() there are only the four strategies (mean, median, most frequent and constant).

python pandas numpy scikit-learn

share|improve this question

asked Mar 24 at 19:59

mweisbrmweisbr

11

• 
  
  
  

  
  1
  

  
  
  
  
  
  

  
  
  What would you like values to be filled with? Your question is a little broad.
  
  – FChm
  Mar 24 at 20:06
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  interpolated values
  
  – mweisbr
  Mar 25 at 7:31
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  Sure, but what method of interpolation do you want?
  
  – FChm
  Mar 25 at 8:06
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  that depends on the results. I have experimented with cubic splines or the Akima method.
  
  – mweisbr
  Mar 25 at 14:50
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  Could you provide your some more context? What if your 2D array? Are the columns strain readings at different positions and the rows readings at different times?
  
  – FChm
  Mar 25 at 15:39
  

  
  
  
  

 | 
show 1 more comment

0

I have very large evaluation data of strain readings which I import as 2d arrays into python. Short Example:

X = np.array([[210, 211, 209, 10000, 215, -5000],
 [220, 221, 219, 10000, 225, -10000],
 1400, 225, 230, 10000, 235, -98]])

Some measured values can be interpreted directly as measurement errors (e.g.:10000,-5000). I define them with an interval (e.g.: <-10, >1000). Then the values should be removed from the array and replaced by suitable values. The measurement errors should either be interpolated or replaced by other methods.

I already had a few attempts with the Panda's function asfreq() and imputation class from scikit-learn. With asfreq() I had the problem that it only fills NaN values. With imputer() there are only the four strategies (mean, median, most frequent and constant).

python pandas numpy scikit-learn

share|improve this question

asked Mar 24 at 19:59

mweisbrmweisbr

11

• 
  
  
  

  
  1
  

  
  
  
  
  
  

  
  
  What would you like values to be filled with? Your question is a little broad.
  
  – FChm
  Mar 24 at 20:06
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  interpolated values
  
  – mweisbr
  Mar 25 at 7:31
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  Sure, but what method of interpolation do you want?
  
  – FChm
  Mar 25 at 8:06
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  that depends on the results. I have experimented with cubic splines or the Akima method.
  
  – mweisbr
  Mar 25 at 14:50
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  Could you provide your some more context? What if your 2D array? Are the columns strain readings at different positions and the rows readings at different times?
  
  – FChm
  Mar 25 at 15:39
  

  
  
  
  

 | 
show 1 more comment

I have very large evaluation data of strain readings which I import as 2d arrays into python. Short Example:

X = np.array([[210, 211, 209, 10000, 215, -5000],
 [220, 221, 219, 10000, 225, -10000],
 1400, 225, 230, 10000, 235, -98]])

Some measured values can be interpreted directly as measurement errors (e.g.:10000,-5000). I define them with an interval (e.g.: <-10, >1000). Then the values should be removed from the array and replaced by suitable values. The measurement errors should either be interpolated or replaced by other methods.

I already had a few attempts with the Panda's function asfreq() and imputation class from scikit-learn. With asfreq() I had the problem that it only fills NaN values. With imputer() there are only the four strategies (mean, median, most frequent and constant).

python pandas numpy scikit-learn

share|improve this question

asked Mar 24 at 19:59

mweisbrmweisbr

11

X = np.array([[210, 211, 209, 10000, 215, -5000],
 [220, 221, 219, 10000, 225, -10000],
 1400, 225, 230, 10000, 235, -98]])

share|improve this question

asked Mar 24 at 19:59

mweisbrmweisbr

11

share|improve this question

asked Mar 24 at 19:59

mweisbrmweisbr

11

asked Mar 24 at 19:59

mweisbrmweisbr

11

asked Mar 24 at 19:59

mweisbrmweisbr

11