Re: home page highlights section updates

[Corston-Oliver, Simon]

+1

I've been meaning to post a digest of the tutorials, blogs, videos posted 
during May but a continually updated feed in a visible place as discussed below 
would be great

On 6/4/18, 9:47 AM, "Lane, Thom"  wrote:

Cheers Aaron. My vote would be for something more like Wireframe 2.

Would be great if we could add thumbnail images to these too, to draw more 
attention and also make the boxes a little less empty.

As Sandeep already mentioned, connecting to our Medium/Twitter/YouTube 
streams would be a great way to get fresh content on the website, but we'd 
probably want an extra category for this. One possibility is;

1) News/Announcements (new versions, apis, toolkits, etc)
2) Social (mostly Twitter, since we tweet about our Medium posts and 
YouTube videos)
3) Tutorials and Examples (combined)

On 6/1/18, 2:58 PM, "Aaron Markham"  wrote:

Hi everyone, happy Friday,

I've mocked up a couple of modifications for the highlights on the home
page.
The goal is to expose more news, more tutorials, and other interesting
things people have been working on. If you've contributed something
interesting and want to highlight it, you should be able to suggest an
update to the page.
It would be great to have variety and fresh content on the site.

Here are a couple of ideas with screenshots that shouldn't be to hard to
implement:

https://cwiki.apache.org/confluence/display/MXNET/Website+Interface+Updates

You can reply with feedback here or make comments on the wiki! What are
your thoughts on:
1) getting fresh content on the site
2) highlighting contributors and their contributions
3) flow on changes and what kind of approvals are expected for swapping 
out
highlights
4) UX ideas that are not too challenging to implement or maintain and 
work
well with mobile

Cheers,
Aaron

Re: multidimensional ndarray indexing issues

[Corston-Oliver, Simon]

The dev email list is mostly used for internal discussion about features etc. 
Could you please post this on the discussion forum: https://discuss.mxnet.io. 
You'll be much more likely to get a response from the community there.

On 2/12/18, 11:24 AM, "Watts 2"  wrote:

I am new to mxnet. I am assuming this is the right place to ask a
programming question. If yes, please see below. If no, please point me to
the right forum.

Thanks

My question -


I just installed mxnet 1.0.0 (later downgraded to 0.12.1 but still same
problem) and python 3.5 on a Ubuntu 14.04 machine with CUDA 8.0 and cudnn
7.0.5.

I am trying to load image data into ndarray.

My code is given below. I am trying to store image data in an ndarray. (see
https://github.com/ypwhs/DogBreed_gluon/blob/master/get_features_v3.ipynb 
for
the original code) -

X_224 = nd.zeros((n, 3, 224, 224))
X_299 = nd.zeros((n, 3, 299, 299))

mean = np.array([0.485, 0.456, 0.406])
std = np.array([0.229, 0.224, 0.225])

for i, (fname, breed) in tqdm(df.iterrows(), total=n):
img = cv2.imread('data/train/%s.jpg' % fname)
img_224 = ((cv2.resize(img, (224, 224))[:, :, ::-1] / 255.0 -
mean) / std).transpose((2, 0, 1))
img_299 = ((cv2.resize(img, (299, 299))[:, :, ::-1] / 255.0 -
mean) / std).transpose((2, 0, 1))

X_224[i] = nd.array(img_224) <-- I get error in this line
X_299[i] = nd.array(img_299)

ValueError: Indexing NDArray with index=0 and type=class 'numpy.int64' is
not supported.