How avoid error "TypeError: invalid data type for einsum" in Python -


i try load csv file numpy-array , use array in logisticregression etc. now, struggling error shown below:

import numpy np import pandas pd  sklearn import preprocessing sklearn.linear_model import logisticregression      dataset =  pd.read_csv('../bookie_test.csv').values x = dataset[1:, 32:34] y = dataset[1:, 14]  # normalize data attributes normalized_x = preprocessing.normalize(x) # standardize data attributes standardized_x = preprocessing.scale(x)  model = logisticregression() model.fit(x, y) print(model) # make predictions expected = y predicted = model.predict(x) # summarize fit of model print(metrics.classification_report(expected, predicted)) print(metrics.confusion_matrix(expected, predicted)) 

i got error:

> c:\anaconda32\lib\site-packages\sklearn\utils\validation.py:332: > userwarning: normalize function assumes floating point values > input, got object   "got %s" % (estimator, x.dtype)) traceback (most > recent call last):   file > "x:/test3.py", line 23, in > <module> >     normalized_x = preprocessing.normalize(x)   file "c:\anaconda32\lib\site-packages\sklearn\preprocessing\data.py", line > 553, in normalize >     norms = row_norms(x)   file "c:\anaconda32\lib\site-packages\sklearn\utils\extmath.py", line 65, > in row_norms >     norms = np.einsum('ij,ij->i', x, x) typeerror: invalid data type einsum 

i new in python , don't transformation:

  1. load csv pandas
  2. convert pandas numpy
  3. use numpy in logisticregression

are there simple approach, like:

  1. load pandas
  2. use pandas dataframes in ml methods?

regarding main question, evert advises check.

regarding #2: found great tutorial http://www.markhneedham.com/blog/2013/11/09/python-making-scikit-learn-and-pandas-play-nice/

and achieved desired result pandas + sklearn


Comments

Popular posts from this blog

PHP DOM loadHTML() method unusual warning -

python - How to create jsonb index using GIN on SQLAlchemy? -

c# - TransactionScope not rolling back although no complete() is called -