there is even a shorter one 🙂 print df.groupby(‘name’).describe().unstack(1) Nothing beats one-liner: In [145]: print df.groupby(‘name’).describe().reset_index().pivot(index=’name’, values=”score”, columns=”level_1″)
Another alternative: df[‘week_start’] = df[‘myday’].dt.to_period(‘W’).apply(lambda r: r.start_time) This will set ‘week_start’ to be the first Monday before the time in ‘myday’. You can choose different week starts via anchored offsets e.g. ’W-THU’ to start the week on Thursday instead. (Thanks @Henry Ecker for that suggestion)
You just need to decode the string back into ASCII, so it would just be: bytes_string.decode(‘UTF-8′) Demo: >>> b’somestring’.decode(‘UTF-8’) ‘somestring’
The value line in each box is telling you how many samples at that node fall into each category, in order. That’s why, in each box, the numbers in value add up to the number shown in sample. For instance, in your red box, 91+212+113=416. So this means if you reach this node, there were … Read more
If I read the source code right (Numpy 1.8.2, Scipy 0.14.1 ), numpy.linalg.lstsq() uses the LAPACK routine xGELSD and scipy.linalg.lstsq() usesxGELSS. The LAPACK Manual Sec. 2.4 states The subroutine xGELSD is significantly faster than its older counterpart xGELSS, especially for large problems, but may require somewhat more workspace depending on the matrix dimensions. That means … Read more
I use Homebrew Python for all my projects (data science, some web dev). Conda is nothing fancy, you can have the same packages by hand with a combination of pip and Homebrew science. Actually, it is even better because you have more control on what you install. You can use your virtualenvs only when you … Read more
You can use dstack: >>> np.dstack((a,b)) array([[[0, 0], [1, 1], [2, 2], [3, 3]], [[4, 4], [5, 5], [6, 6], [7, 7]]]) If you must have tuples: >>> np.array(zip(a.ravel(),b.ravel()), dtype=(‘i4,i4’)).reshape(a.shape) array([[(0, 0), (1, 1), (2, 2), (3, 3)], [(4, 4), (5, 5), (6, 6), (7, 7)]], dtype=[(‘f0’, ‘<i4’), (‘f1’, ‘<i4’)]) For Python 3+ you need … Read more
numpy.show_config() doesn’t always give reliable information. For example, if I apt-get install python-numpy on Ubuntu 14.04, the output of np.show_config() looks like this: blas_info: libraries = [‘blas’] library_dirs = [‘/usr/lib’] language = f77 lapack_info: libraries = [‘lapack’] library_dirs = [‘/usr/lib’] language = f77 atlas_threads_info: NOT AVAILABLE blas_opt_info: libraries = [‘blas’] library_dirs = [‘/usr/lib’] language = … Read more
Try concatenating X_Yscores[:, None] (or X_Yscores[:, np.newaxis] as imaluengo suggests). This creates a 2D array out of a 1D array. Example: A = np.array([1, 2, 3]) print A.shape print A[:, None].shape Output: (3,) (3,1)
This might not be obvious, but pd.Series.isin uses O(1)-look up per element. After an analysis, which proves the above statement, we will use its insights to create a Cython-prototype which can easily beat the fastest out-of-the-box-solution. Let’s assume that the “set” has n elements and the “series” has m elements. The running time is then: … Read more