Skip to content

Fix weighted sampling without replacement #239

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 3 commits into from
May 4, 2017
Merged

Fix weighted sampling without replacement #239

Changes from 1 commit
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
60cbcc9
Fix #238
bkamins Feb 15, 2017
5ed66f3
weighted sampling without replacement: added tests and small fixes
bkamins Feb 16, 2017
77ec83a
removed redundant import in tests
bkamins May 3, 2017
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
27 changes: 21 additions & 6 deletions src/sampling.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -515,15 +515,22 @@ function efraimidis_ares_wsample_norep!(a::AbstractArray, wv::WeightVec, x::Abst

# initialize priority queue
pq = Vector{Pair{Float64,Int}}(k)
@inbounds for i in 1:k
pq[i] = (wv.values[i]/randexp() => i)
i = 0
s = 0
@inbounds for s in 1:n
if wv.values[s] > 0.0
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

> 0 should be enough, no?

i += 1
pq[i] = (wv.values[s]/randexp() => s)
end
i >= k && break
end
i < k && throw(DimensionMismatch("wv must have at least $k positive entries (got $i)"))
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

"strictly positive"

heapify!(pq)

# set threshold
@inbounds threshold = pq[1].first

@inbounds for i in k+1:n
@inbounds for i in s+1:n
key = wv.values[i]/randexp()

# if key is larger than the threshold
Expand Down Expand Up @@ -561,17 +568,25 @@ function efraimidis_aexpj_wsample_norep!(a::AbstractArray, wv::WeightVec, x::Abs

# initialize priority queue
pq = Vector{Pair{Float64,Int}}(k)
@inbounds for i in 1:k
pq[i] = (wv.values[i]/randexp() => i)
i = 0
s = 0
@inbounds for s in 1:n
if wv.values[s] > 0.0
i += 1
pq[i] = (wv.values[s]/randexp() => s)
end
i >= k && break
end
i < k && throw(DimensionMismatch("wv must have at least $k positive entries (got $i)"))
heapify!(pq)

# set threshold
@inbounds threshold = pq[1].first
X = threshold*randexp()

@inbounds for i in k+1:n
@inbounds for i in s+1:n
w = wv.values[i]
w > 0.0 || continue
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Same here. Would be more consistent to use > 0 since <= 0 is used below.

X -= w
X <= 0 || continue

Expand Down