Skip to content

Instantly share code, notes, and snippets.

@rjurney rjurney/
Created Oct 22, 2019

What would you like to do?
Custom padding of dense vectors with min/max or mean
padded_posts = []
for post in encoded_docs:
# Pad short posts with alternating min/max
if len(post) < MAX_LENGTH:
# Method 1
pointwise_min = np.minimum.reduce(post)
pointwise_max = np.maximum.reduce(post)
padding = [pointwise_max, pointwise_min]
# Method 2
pointwise_avg = np.mean(post)
padding = [pointwise_avg]
post += padding * ceil((MAX_LENGTH - len(post) / 2.0))
# Shorten long posts or those odd number length posts we padded to 51
if len(post) > MAX_LENGTH:
post = post[:MAX_LENGTH]
# Add the post to our new list of padded posts
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
You can’t perform that action at this time.