Anaconda Cloud for MXNet being maintained?


#1
(mxnet2) user@GPU2:~/$ conda install -c anaconda mxnet-gpu  // similarly for just mxnet
Solving environment: done


==> WARNING: A newer version of conda exists. <==
  current version: 4.4.10
  latest version: 4.5.11

Please update conda by running

    $ conda update -n base conda



## Package Plan ##

  environment location: /usr/local/Anaconda3-5.1.0-Linux-x86_64/envs/mxnet2

  added / updated specs:
    - mxnet-gpu


The following packages will be downloaded:

    package                    |            build
    ---------------------------|-----------------
    mxnet-1.2.1                |       h8cc8929_0           3 KB  anaconda

The following NEW packages will be INSTALLED:

    _mutex_mxnet:         0.0.20-gpu_mkl           anaconda
    asn1crypto:           0.24.0-py36_0            anaconda
    blas:                 1.0-mkl                  anaconda
    bzip2:                1.0.6-h14c3975_5         anaconda
    ca-certificates:      2018.03.07-0             anaconda
    cairo:                1.14.12-h8948797_3       anaconda
    certifi:              2018.8.24-py36_1         anaconda
    cffi:                 1.11.5-py36he75722e_1    anaconda
    chardet:              3.0.4-py36_1             anaconda
    cryptography:         2.3-py36hb7f436b_1       anaconda
    cryptography-vectors: 2.3.1-py36_0             anaconda
    cudatoolkit:          8.0-3                    anaconda
    cudnn:                7.1.3-cuda8.0_0          anaconda
    ffmpeg:               3.4-h7985aa0_0           anaconda
    fontconfig:           2.13.0-h9420a91_0        anaconda
    freeglut:             3.0.0-hf484d3e_5         anaconda
    freetype:             2.9.1-h8a8886c_0         anaconda
    glib:                 2.56.1-h000015b_0        anaconda
    graphite2:            1.3.11-h16798f4_2        anaconda
    harfbuzz:             1.8.4-hec2c2bc_0         anaconda
    hdf5:                 1.10.2-hba1933b_1        anaconda
    icu:                  58.2-h211956c_0          anaconda
    idna:                 2.6-py36h82fb2a8_1       anaconda
    intel-openmp:         2018.0.3-0               anaconda
    jasper:               2.0.14-h07fcdf6_1        anaconda
    jpeg:                 9b-h024ee3a_2            anaconda
    libedit:              3.1.20170329-h6b74fdf_2  anaconda
    libffi:               3.2.1-h4deb6c0_3         anaconda
    libgcc-ng:            7.2.0-hdf63c60_3         anaconda
    libgfortran-ng:       7.3.0-hdf63c60_0         anaconda
    libglu:               9.0.0-hf484d3e_1         anaconda
    libmklml:             2018.0.3-0               anaconda
    libmxnet:             1.2.1-gpu_mkl_h3d71631_1 anaconda
    libopencv:            3.4.1-h8fa1ad8_3         anaconda
    libopus:              1.2.1-hb9ed12e_0         anaconda
    libpng:               1.6.34-hb9fc6fc_0        anaconda
    libprotobuf:          3.5.2-h6f1eeef_0         anaconda
    libstdcxx-ng:         8.2.0-hdf63c60_1         anaconda
    libtiff:              4.0.9-he85c1e1_1         anaconda
    libuuid:              1.0.3-h1bed415_2         anaconda
    libvpx:               1.7.0-h439df22_0         anaconda
    libxcb:               1.13-h1bed415_1          anaconda
    libxml2:              2.9.8-h26e45fe_1         anaconda
    mkl:                  2019.0-118               anaconda
    mkl-dnn:              0.14-h6bb024c_0          anaconda
    mkl_fft:              1.0.4-py36h4414c95_1     anaconda
    mkl_random:           1.0.1-py36h4414c95_1     anaconda
    mxnet:                1.2.1-h8cc8929_0         anaconda
    mxnet-gpu:            1.2.1-hf82a2c8_0         anaconda
    ncurses:              6.1-hf484d3e_0           anaconda
    numpy:                1.15.0-py36h1b885b7_0    anaconda
    numpy-base:           1.15.0-py36h3dfced4_0    anaconda
    openssl:              1.0.2p-h14c3975_0        anaconda
    pcre:                 8.42-h439df22_0          anaconda
    pip:                  10.0.1-py36_0            anaconda
    pixman:               0.34.0-hceecf20_3        anaconda
    py-mxnet:             1.2.1-py36h6c82189_0     anaconda
    pycparser:            2.18-py36_1              anaconda
    pyopenssl:            18.0.0-py36_0            anaconda
    pysocks:              1.6.8-py36_0             anaconda
    python:               3.6.6-hc3d631a_0         anaconda
    readline:             7.0-ha6073c6_4           anaconda
    requests:             2.18.4-py36he2e5f8d_1    anaconda
    setuptools:           40.2.0-py36_0            anaconda
    six:                  1.11.0-py36_1            anaconda
    sqlite:               3.24.0-h84994c4_0        anaconda
    tk:                   8.6.7-hc745277_3         anaconda
    urllib3:              1.22-py36hbe7ace6_0      anaconda
    wheel:                0.31.1-py36_0            anaconda
    xz:                   5.2.4-h14c3975_4         anaconda
    zlib:                 1.2.11-ha838bed_2        anaconda

Proceed ([y]/n)? y


Downloading and Extracting Packages
mxnet 1.2.1:                                                             |   0%

CondaHTTPError: HTTP 404 NOT FOUND for url <https://conda.anaconda.org/anaconda/linux-64/mxnet-1.2.1-h8cc8929_0.tar.bz2>
Elapsed: 00:00.416062
CF-RAY: 46319665caeb590f-DFW

An HTTP error occurred when trying to retrieve this URL.
HTTP errors are often intermittent, and a simple retry will get you on your way.


https://anaconda.org/anaconda/mxnet-gpu/files shows that linux have 1.2.1 support; however, it is only 1.1.0 as shown here: https://anaconda.org/anaconda/mxnet/files?version=1.2.1


#2

Hi @lyandrew,

It looks as if these packages are added by Anaconda. I see quite a bit of past activity from Kale Franz who’s a lead developer at Anaconda (https://twitter.com/kalefranz), so it might be worth reaching out to him for more details. MXNet releases are officially provided through pypi (https://pypi.org/project/mxnet/), docker hub (https://hub.docker.com/r/mxnet/python/) and also apache (https://www.apache.org/dyn/closer.cgi/incubator/mxnet).

Similar conda packages you might want to check out are https://anaconda.org/anaconda/libmxnet and https://anaconda.org/anaconda/py-mxnet which both look like they have linux versions for v1.2.1.