Skip to content

Instantly share code, notes, and snippets.

@vincentchu

vincentchu/ssd_full_infer.ipynb

Created January 23, 2018 18:00
Show Gist options
  • Star 5 You must be signed in to star a gist
  • Fork 1 You must be signed in to fork a gist
  • Save vincentchu/5f2f669aeb6df9864ec6c43252261269 to your computer and use it in GitHub Desktop.
Save vincentchu/5f2f669aeb6df9864ec6c43252261269 to your computer and use it in GitHub Desktop.
Download ZIP
Display the source blob
Display the rendered blob
Raw
ssd_full_infer.ipynb
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
@dhasl002
Copy link

dhasl002 commented Jun 30, 2018 •
edited

I think this line is hard coded:

"sorted_high_scores = np.concatenate((high_scoring_scores[sorted_inds].reshape(7, 1), high_scoring_boxes, high_scoring_anchors), axis=1)"

Changing it to this would be better:

"sorted_high_scores = np.concatenate((high_scoring_scores[sorted_inds].reshape(high_scoring_scores.shape[0], 1), high_scoring_boxes, high_scoring_anchors), axis=1)"

@tmyapple
Copy link

first of all thanks for this git, it is very helpful ...

I have a question about what you did in ln[6]
I get the following error:
InvalidArgumentError (see above for traceback): You must feed a value for placeholder tensor 'image_tensor' with dtype uint8 and shape [?,?,?,3]
-- from my experience tensorflow cannot run the graph without any input given to the placeholders, so even-though there's the "Preprocessor/sub:0", it is not enough because the post processing block needs the image_tensor (for dimensions) - I fixed it by feeding a dummy image to the placeholder...

What tensorflow version are you using ? maybe the error I got is because of version differences?
Just trying to figure out what I'm missing here...

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment