CODEWITHSUNDEEP
python-2.7scikit-learnpca
user122508
user122508

Reputation: 15

Python float error with numerical data

I'm trying to run PCA using sklearn on a dataset with 162 columns and 69,000 rows. I keep getting the float error message below and I've checked to make sure I only have numerical data. What could I be doing wrong? Any help would be really appreciated.

    >>> data = np.loadtxt("PCAdata.txt")
    >>> trans = data.transpose()
    >>> trans
    array([[0., 0., 1., ..., 0., 0., 1.],
           [0., 0., 1., ..., 1., 0., 2.],
           [0., 0., 1., ..., 0., 0., 1.],
           ...,
           [1., 0., 1., ..., 0., 0., 1.],
           [0., 0., 1., ..., 0., 0., 2.],
           [0., 0., 1., ..., 0., 0., 2.]])
    >>> sscaler = preprocessing.StandardScaler().fit(trans)
    >>> sscaler
    StandardScaler(copy=True, with_mean=True, with_std=True)
    >>> pca = PCA(n_components=2)
    >>> pca.fit(sscaler)
    Traceback (most recent call last):
      File "<stdin>", line 1, in <module>
      File "C:\Python27\lib\site-packages\sklearn\decomposition\pca.py", line 329, i
    n fit
        self._fit(X)
      File "C:\Python27\lib\site-packages\sklearn\decomposition\pca.py", line 370, i
    n _fit
        copy=self.copy)
      File "C:\Python27\lib\site-packages\sklearn\utils\validation.py", line 433, in
     check_array
        array = np.array(array, dtype=dtype, order=order, copy=copy)
    TypeError: float() argument must be a string or a number

Upvotes: 0

Views: 378

Answers (1)

Seljuk Gulcan
Seljuk Gulcan

Reputation: 1868

fit method does not return a matrix. Sklearn gives error because parameter you feed, sscaler, is not a matrix of numbers. If you want to get scaled data matrix you may use fit_transform method or use fit and transform methods separately.

Example :

data = np.random.randint(0, 3, (100, 10))
scaler = StandardScaler()
data = scaler.fit_transform(data)
pca = PCA()
data = pca.fit_transform(data)

Upvotes: 1

Related Questions