Skip to content
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

Jump to bottom

float64 support in FIL core #4646

Merged
merged 51 commits into from
Apr 7, 2022
Merged

float64 support in FIL core #4646

merged 51 commits into from
Apr 7, 2022

Conversation

canonizer
Copy link
Contributor

@canonizer canonizer commented Mar 22, 2022 •
edited
Loading

float64 support in FIL core and tests.

Added float64-based instantiations of FIL types and functions, together with float64-based FIL tests.

Treelite and Python layers are not covered here, but will be in separate pull requests.

This is pull request 3 of 3 to enable float64 support in FIL core. This is partly based on the work by @levsnv.

levsnv and others added 30 commits February 8, 2022 19:59
@levsnv
templatized node, forest and storage types
86bc6b5
@levsnv
compiles?
abfb605
@levsnv
Merge branch 'branch-22.04' of github.com:rapidsai/cuml into fp64
a5d970f
@levsnv
simplify leaf_output_t
96e2ae6
@levsnv
Merge branch 'branch-22.04' of github.com:rapidsai/cuml into fp64
ead832d
@levsnv
draft
35ad5d9
@levsnv
Merge branch 'branch-22.04' of github.com:rapidsai/cuml into fp64
df001e7
@levsnv
fixed extra/missing instantiations
f6efb88
@levsnv
Merge branch 'fp64' into func64
4bc4f99
@levsnv
style
1bebcca
@levsnv
style
aab158b
@levsnv
Merge branch 'fp64' into func64
c3f2d4f
@levsnv
removed ML::fil::init templatization, added KeyValuePair templatization
1236c52
@levsnv
style
cc946cd
@levsnv
add the old instantiations back
91ecb20
@levsnv
fixed
67f900e
@levsnv
style
c3c4deb
@canonizer
base_node::output() now compiles.
68e7f65
@canonizer
Merge branch 'branch-22.04' into dev-fil-fp64
fe2cf92
@canonizer
Fixed style.
b31069d
@canonizer
F -> real_t.
c250653
@canonizer
Small fixes.
2f658b5
@canonizer
Updated alignment.
6fb105b
@canonizer
static_assert(real_t == float) in a number of places.
f1a10be
@canonizer
noinline -> forceinline.
ce3624e
@canonizer
Updated comment.
63fadd1
@canonizer
Merge branch 'dev-fil-fp64' into enh-fil-func64
986a50d
@canonizer
Fixed many compiler errors.
fda7669
@canonizer
Multiple changes.
40f7a23
@canonizer
Fixed compilation errors; now it compiles.
12cc051
@github-actions github-actions bot removed CMake conda conda issue gpuCI gpuCI issue Cython / Python Cython or Python issue labels Apr 4, 2022
@canonizer canonizer changed the title [WIP] [skip ci] float64 support in FIL core float64 support in FIL core Apr 4, 2022
@canonizer
Copy link
Contributor Author

rerun tests

divyegala
divyegala reviewed Apr 6, 2022
View reviewed changes
Copy link
Member

@divyegala divyegala left a comment

Choose a reason for hiding this comment

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

Just a couple of comments.

cpp/test/sg/fil_test.cu Outdated Show resolved Hide resolved
cpp/test/sg/fil_test.cu Show resolved Hide resolved
@canonizer
Copy link
Contributor Author

@divyegala Thanks for your comments! I've addressed them, could you take another look?

divyegala
divyegala approved these changes Apr 6, 2022
View reviewed changes
Copy link
Member

@divyegala divyegala left a comment

Choose a reason for hiding this comment

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

LGTM!

@canonizer canonizer mentioned this pull request Apr 7, 2022
Merged
@canonizer
Copy link
Contributor Author

rerun tests

@dantegd dantegd added feature request New feature or request non-breaking Non-breaking change labels Apr 7, 2022
@codecov-commenter
Copy link

Codecov Report

❗ No coverage uploaded for pull request base (branch-22.06@c4c80aa). Click here to learn what that means.
The diff coverage is n/a.

@@               Coverage Diff               @@
##             branch-22.06    #4646   +/-   ##
===============================================
  Coverage                ?   84.06%           
===============================================
  Files                   ?      252           
  Lines                   ?    20336           
  Branches                ?        0           
===============================================
  Hits                    ?    17095           
  Misses                  ?     3241           
  Partials                ?        0
Flag Coverage Δ
dask 44.98% <0.00%> (?)
non-dask 77.35% <0.00%> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update c4c80aa...56ecd51. Read the comment docs.

@dantegd
Copy link
Member

dantegd commented Apr 7, 2022

@gpucibot merge

@rapids-bot rapids-bot bot merged commit be49919 into rapidsai:branch-22.06 Apr 7, 2022
venkywonka pushed a commit to venkywonka/cuml that referenced this pull request Apr 26, 2022
@canonizer @venkywonka
float64 support in multi-sum and child_index() (rapidsai#4648)
28a027e 
- removed default `T = float` from multi-sum
- tests with float64 for multi-sum and `child_index()`
- refactored multi-sum tests to reduce shared memory usage
  - this is needed for the tests with float64 to compile

This is pull request 1 of 3 to integrate rapidsai#4646. This pull request is partly based on the work by @levsnv.

Authors:
  - Andy Adinets (https://github.com/canonizer)

Approvers:
  - Divye Gala (https://github.com/divyegala)
  - William Hicks (https://github.com/wphicks)

URL: rapidsai#4648
venkywonka pushed a commit to venkywonka/cuml that referenced this pull request Apr 26, 2022
@canonizer @venkywonka
float64 support in FIL functions (rapidsai#4655)
a3ea64d 
Templatized functions related to FIL inference in preparation of `float64` support.

Instantiations of templates with `float64`, or tests for `float64`, _are not included_; they will be provided in a future pull request.

This is pull request 2 of 3 to integrate rapidsai#4646. This pull request is partly based on the work by @levsnv.

Authors:
  - Andy Adinets (https://github.com/canonizer)
  - Levs Dolgovs (https://github.com/levsnv)
  - Dante Gama Dessavre (https://github.com/dantegd)

Approvers:
  - Divye Gala (https://github.com/divyegala)
  - William Hicks (https://github.com/wphicks)

URL: rapidsai#4655
venkywonka pushed a commit to venkywonka/cuml that referenced this pull request Apr 26, 2022
@canonizer @venkywonka
float64 support in FIL core (rapidsai#4646)
d2f64f5 
`float64` support in FIL core and tests.

Added `float64`-based instantiations of FIL types and functions, together with `float64`-based FIL tests.

Treelite and Python layers _are not covered_ here, but will be in separate pull requests.

This is pull request 3 of 3 to enable `float64` support in FIL core. This is partly based on the work by @levsnv.

Authors:
  - Andy Adinets (https://github.com/canonizer)
  - Levs Dolgovs (https://github.com/levsnv)

Approvers:
  - Divye Gala (https://github.com/divyegala)

URL: rapidsai#4646
vimarsh6739 pushed a commit to vimarsh6739/cuml that referenced this pull request Oct 9, 2023
@canonizer
float64 support in multi-sum and child_index() (rapidsai#4648)
29057cb 
- removed default `T = float` from multi-sum
- tests with float64 for multi-sum and `child_index()`
- refactored multi-sum tests to reduce shared memory usage
  - this is needed for the tests with float64 to compile

This is pull request 1 of 3 to integrate rapidsai#4646. This pull request is partly based on the work by @levsnv.

Authors:
  - Andy Adinets (https://github.com/canonizer)

Approvers:
  - Divye Gala (https://github.com/divyegala)
  - William Hicks (https://github.com/wphicks)

URL: rapidsai#4648
vimarsh6739 pushed a commit to vimarsh6739/cuml that referenced this pull request Oct 9, 2023
@canonizer
float64 support in FIL functions (rapidsai#4655)
3ce72e6 
Templatized functions related to FIL inference in preparation of `float64` support.

Instantiations of templates with `float64`, or tests for `float64`, _are not included_; they will be provided in a future pull request.

This is pull request 2 of 3 to integrate rapidsai#4646. This pull request is partly based on the work by @levsnv.

Authors:
  - Andy Adinets (https://github.com/canonizer)
  - Levs Dolgovs (https://github.com/levsnv)
  - Dante Gama Dessavre (https://github.com/dantegd)

Approvers:
  - Divye Gala (https://github.com/divyegala)
  - William Hicks (https://github.com/wphicks)

URL: rapidsai#4655
vimarsh6739 pushed a commit to vimarsh6739/cuml that referenced this pull request Oct 9, 2023
@canonizer
float64 support in FIL core (rapidsai#4646)
64c7f60 
`float64` support in FIL core and tests.

Added `float64`-based instantiations of FIL types and functions, together with `float64`-based FIL tests.

Treelite and Python layers _are not covered_ here, but will be in separate pull requests.

This is pull request 3 of 3 to enable `float64` support in FIL core. This is partly based on the work by @levsnv.

Authors:
  - Andy Adinets (https://github.com/canonizer)
  - Levs Dolgovs (https://github.com/levsnv)

Approvers:
  - Divye Gala (https://github.com/divyegala)

URL: rapidsai#4646
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@divyegala divyegala divyegala approved these changes

Assignees
No one assigned
Labels
CUDA/C++ feature request New feature or request non-breaking Non-breaking change
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
6 participants
@canonizer @ajschmidt8 @codecov-commenter @dantegd @divyegala @levsnv