python - sci-kit learn crashing on certain amounts of data -


I am trying to process 71,000 rows of more than 200 columns with a lower array and two science-kit models I learn I'm trying to give both different error when I tried to remove the problem line more than 5853 rows, but it failed. Can not science-kit learn to handle this much data, or is it something else? A list of XLists is a narrow array.

KNN:

  nbrs = Nearest navy (n_neighbors = 2, algorithm = 'ball_tree'). Fit (x)  

Error:

  File "knn.py", line 48,  

Value error: data type is not understood

K-mean:

  kmeans_model = KMines (n_clusters = 2, random_state = 1). (X)  

Error:

  Traceback (last call final): File "knn.py", line 48,  

Please check dtype of your matrix X By typing X.dtype , if it is object or dtype ('o') , then in an array

  length = [for line in x] [lane (line)]  

then it Take a look to see if all the rows are of the same length, using

  np.unique (length)  

In more than one numbers, your line length is different, eg. From line 5853, probably not at all times.

Numbers data arrays are useful only when all lines have the same length (they continue to work, but do not do what you expect.) You should check to see why this is the reason, fix it, and then return to knn .

Here is an example, what would happen if the line length is not equal:

  np rng = np.random.RandomState (42) x = rng.randn ( 100, 20) Import samples as # Now remove an element from the 56th line X = list (X) X [55] = x [55] [: - 1] # back it ndarray x = np Change in .array (X) # Change dtype from X.dtype # sklearn.neighbors Returns dtype ('O') nearest young nbrs =  
/ Div>
html>

Comments

Popular posts from this blog

import - Python ImportError: No module named wmi -

Editing Python Class in Shell and SQLAlchemy -

c# - MySQL Parameterized Select Query joining tables issue -