-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 Dear all, Prerelease binary wheels for Scipy on Windows 32-bit & 64-bit are now available in case you would like to test them. Currently, the plans are that binary wheels will also be provided for future releases on PyPi, so that you will be able to do simply "pip install scipy" also on Windows. At least, assuming we manage to test these wheels well enough for which help would be useful. You can install the scipy prerelease packages as shown below. Note that they are meant for testing only, and correspond to the current Scipy development version. Please report issues found on the Scipy issue tracker on github (be sure to mention how you installed scipy and python). The wheels are meant to be used with the Python obtained from https://p ython.org --- these are not meant to be used with e.g. Conda, although it may be they work. The work leading to a viable automatized compilation approach was done in https://github.com/scipy/scipy/pull/7616 https://github.com/numpy/numpy/pull/9431 Example: C:\Users\pauli\src\env2\Scripts>pip install -f https://7933911d6844c6c53a7d-47bd50c35cd79bd838daf386af554a83.ssl.cf2.rackcd... --pre scipy Collecting scipy Downloading https://7933911d6844c6c53a7d-47bd50c35cd79bd838daf386af554a83.ssl.cf2.rackcd... (26.0MB) 100% |████████████████████████████████| 26.0MB 47kB/s Collecting numpy>=1.8.2 (from scipy) Downloading https://7933911d6844c6c53a7d-47bd50c35cd79bd838daf386af554a83.ssl.cf2.rackcd... (6.8MB) 100% |████████████████████████████████| 6.9MB 168kB/s Installing collected packages: numpy, scipy Successfully installed numpy-1.14.0.dev0+707f33f scipy-1.0.0.dev0+2a1fdcf C:\Users\pauli\src\env2\Scripts>python Python 3.6.2 (v3.6.2:5fd33b5, Jul 8 2017, 04:14:34) [MSC v.1900 32 bit (Intel)] on win32 Type "help", "copyright", "credits" or "license" for more information.
import scipy.integrate, scipy.linalg, numpy as np scipy.integrate.quad(lambda x: 1/(1 + x**2), -np.inf, np.inf) (3.141592653589793, 5.155583041103855e-10) scipy.linalg.eigvals([[1,0],[0,2]]) array([ 1.+0.j, 2.+0.j]) exit()
- -- Pauli Virtanen -----BEGIN PGP SIGNATURE----- iQIzBAEBCAAdFiEEiNDtwDFNJaodBiJZHOiglY3YpVcFAlmfWaUACgkQHOiglY3Y pVfxSg//VvwYDanA/tam34I2hrwQQFtQaW3E/ikqNAM91XvQwluc9RsVVDx2zspk 9ywXyfFbnNpMK+/emow8o6kuaFes1NvutobbAOQ4L9jj0ofD9pCWVM6SLkRIkSea km7RdK13vpWtrghPkvkqGFNhY2eDSuV4S8qeR+78KwSUADYjB0m1Yfpfm6LtUKOX tKwGhDGWzi1vcBPJqgQQYJDjBbVNbY5aao6QnjLeNkgXW6RZYhxUyBeWph7GPrEL pNFvoOknjxa5nItZvt948+7PsgZZarHGlyqeAy8Nb0Bkukm1Uovo7V5gMfiDS6nT cA2xNkELh9Zoyr+9kRaaDh2B0U6qWPOiU/IE6VvCK72N70tzdi59a0GkzLzVen2b hgK5RBsa5fL9sxo4oN/bcApnUp6K98XAV4eJhIlZPbnnvSfqbKobX7D1G+qBokBN 90XnWLUjkJpzr1emqyrQPVbrd8OflIhs2aQv0l5gZKrXuBgGFgoCDwEJmrzd6K+n 1iLr73BuZEFN/jLvT9cx+XbAQkXhCbD2hL4ly0u7BuBzAbOE19iugSnap/sjueRW FlOKSddobW86TeOICKurH9TCcFRu6mu1tQvCkucqkY49gXpu3srzUcdog9gQe46H 2JFNQICFaYWhF7jVY9cwOXssEHOc6PCa0FdOxMX5W/p5k0xuqzA= =tjJg -----END PGP SIGNATURE-----
On Thu, Aug 24, 2017 at 6:56 PM, Pauli Virtanen <pav@iki.fi> wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256
Dear all,
Prerelease binary wheels for Scipy on Windows 32-bit & 64-bit are now available in case you would like to test them.
Currently, the plans are that binary wheels will also be provided for future releases on PyPi, so that you will be able to do simply "pip install scipy" also on Windows. At least, assuming we manage to test these wheels well enough for which help would be useful.
You can install the scipy prerelease packages as shown below. Note that they are meant for testing only, and correspond to the current Scipy development version. Please report issues found on the Scipy issue tracker on github (be sure to mention how you installed scipy and python).
The wheels are meant to be used with the Python obtained from https://p ython.org --- these are not meant to be used with e.g. Conda, although it may be they work.
What's the numpy requirement? I assume the scipy version should not be used with a currently installed numpy unless it is Fortran compatible. For example, Winpython distributes Gohlke's binaries build with MKL. Is there an automatic check when not installing into an empty virtual environment? Josef
The work leading to a viable automatized compilation approach was done in https://github.com/scipy/scipy/pull/7616 https://github.com/numpy/numpy/pull/9431
Example:
C:\Users\pauli\src\env2\Scripts>pip install -f https://7933911d6844c6c53a7d-47bd50c35cd79bd838daf386af554a 83.ssl.cf2.rackcdn.com/ --pre scipy Collecting scipy Downloading https://7933911d6844c6c53a7d-47bd50c35cd79bd838daf386af554a 83.ssl.cf2.rackcdn.com/scipy-1.0.0.dev0+20170824221943_ 2a1fdcf-cp36-none-win32.whl (26.0MB) 100% |████████████████████████████████| 26.0MB 47kB/s Collecting numpy>=1.8.2 (from scipy) Downloading https://7933911d6844c6c53a7d-47bd50c35cd79bd838daf386af554a 83.ssl.cf2.rackcdn.com/numpy-1.14.0.dev0+20170824081646_ 707f33f-cp36-none-win32.whl (6.8MB) 100% |████████████████████████████████| 6.9MB 168kB/s Installing collected packages: numpy, scipy Successfully installed numpy-1.14.0.dev0+707f33f scipy-1.0.0.dev0+2a1fdcf
C:\Users\pauli\src\env2\Scripts>python Python 3.6.2 (v3.6.2:5fd33b5, Jul 8 2017, 04:14:34) [MSC v.1900 32 bit (Intel)] on win32 Type "help", "copyright", "credits" or "license" for more information.
import scipy.integrate, scipy.linalg, numpy as np scipy.integrate.quad(lambda x: 1/(1 + x**2), -np.inf, np.inf) (3.141592653589793, 5.155583041103855e-10) scipy.linalg.eigvals([[1,0],[0,2]]) array([ 1.+0.j, 2.+0.j]) exit()
- -- Pauli Virtanen -----BEGIN PGP SIGNATURE-----
iQIzBAEBCAAdFiEEiNDtwDFNJaodBiJZHOiglY3YpVcFAlmfWaUACgkQHOiglY3Y pVfxSg//VvwYDanA/tam34I2hrwQQFtQaW3E/ikqNAM91XvQwluc9RsVVDx2zspk 9ywXyfFbnNpMK+/emow8o6kuaFes1NvutobbAOQ4L9jj0ofD9pCWVM6SLkRIkSea km7RdK13vpWtrghPkvkqGFNhY2eDSuV4S8qeR+78KwSUADYjB0m1Yfpfm6LtUKOX tKwGhDGWzi1vcBPJqgQQYJDjBbVNbY5aao6QnjLeNkgXW6RZYhxUyBeWph7GPrEL pNFvoOknjxa5nItZvt948+7PsgZZarHGlyqeAy8Nb0Bkukm1Uovo7V5gMfiDS6nT cA2xNkELh9Zoyr+9kRaaDh2B0U6qWPOiU/IE6VvCK72N70tzdi59a0GkzLzVen2b hgK5RBsa5fL9sxo4oN/bcApnUp6K98XAV4eJhIlZPbnnvSfqbKobX7D1G+qBokBN 90XnWLUjkJpzr1emqyrQPVbrd8OflIhs2aQv0l5gZKrXuBgGFgoCDwEJmrzd6K+n 1iLr73BuZEFN/jLvT9cx+XbAQkXhCbD2hL4ly0u7BuBzAbOE19iugSnap/sjueRW FlOKSddobW86TeOICKurH9TCcFRu6mu1tQvCkucqkY49gXpu3srzUcdog9gQe46H 2JFNQICFaYWhF7jVY9cwOXssEHOc6PCa0FdOxMX5W/p5k0xuqzA= =tjJg -----END PGP SIGNATURE-----
_______________________________________________ SciPy-Dev mailing list SciPy-Dev@python.org https://mail.python.org/mailman/listinfo/scipy-dev
On Thu, Aug 24, 2017 at 4:49 PM, <josef.pktd@gmail.com> wrote:
On Thu, Aug 24, 2017 at 6:56 PM, Pauli Virtanen <pav@iki.fi> wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256
Dear all,
Prerelease binary wheels for Scipy on Windows 32-bit & 64-bit are now available in case you would like to test them.
Currently, the plans are that binary wheels will also be provided for future releases on PyPi, so that you will be able to do simply "pip install scipy" also on Windows. At least, assuming we manage to test these wheels well enough for which help would be useful.
You can install the scipy prerelease packages as shown below. Note that they are meant for testing only, and correspond to the current Scipy development version. Please report issues found on the Scipy issue tracker on github (be sure to mention how you installed scipy and python).
The wheels are meant to be used with the Python obtained from https://p ython.org --- these are not meant to be used with e.g. Conda, although it may be they work.
What's the numpy requirement? I assume the scipy version should not be used with a currently installed numpy unless it is Fortran compatible. For example, Winpython distributes Gohlke's binaries build with MKL. Is there an automatic check when not installing into an empty virtual environment?
IIUC the only reason Gohlke's scipy builds require a specific numpy is that they use a tricky hack where they assume the numpy package has installed MKL at a specific location. I don't think these builds use any hack like this and should work with any numpy. (This makes the download size larger b/c it means scipy has to carry its own copy of BLAS, but the trade-off in "just works" is worth it IMO.) Also, numpy itself doesn't provide any Fortran APIs, so Fortran ABI shouldn't matter at all. -n -- Nathaniel J. Smith -- https://vorpus.org
to, 2017-08-24 kello 16:58 -0700, Nathaniel Smith kirjoitti: [clip]
IIUC the only reason Gohlke's scipy builds require a specific numpy is that they use a tricky hack where they assume the numpy package has installed MKL at a specific location. I don't think these builds use any hack like this and should work with any numpy. (This makes the download size larger b/c it means scipy has to carry its own copy of BLAS, but the trade-off in "just works" is worth it IMO.)
Also, numpy itself doesn't provide any Fortran APIs, so Fortran ABI shouldn't matter at all.
These wheels lug their own Openblas and gfortran libs with them, so in theory at least they should only care about the correct Numpy version, and the correct Python CRT. Pauli
pe, 2017-08-25 kello 02:08 +0200, Pauli Virtanen kirjoitti: [clip]
These wheels lug their own Openblas and gfortran libs with them, so in theory at least they should only care about the correct Numpy version, and the correct Python CRT.
Indeed, they appear to work fine also in practice with Gohlke's numpy package. Pauli
to, 2017-08-24 kello 19:49 -0400, josef.pktd@gmail.com kirjoitti: [clip]
What's the numpy requirement? I assume the scipy version should not be used with a currently installed numpy unless it is Fortran compatible. For example, Winpython distributes Gohlke's binaries build with MKL. Is there an automatic check when not installing into an empty virtual environment?
Any official Numpy wheel available on Pypi should be fine. You get ABI version errors at import time as usual if you get it wrong. If winpython installs a python that uses different CRT than the python.org ones, you may get extra problems. Pauli
participants (3)
-
josef.pktd＠gmail.com -
Nathaniel Smith -
Pauli Virtanen