gpu-computing
Here are 431 public repositories matching this topic...
-
Updated
Jan 19, 2021 - Python
Heston model has accurate density approximations for European option prices, which are of interest.
The module implementing this method should live under tf_quant_finance/volatility/heston_approximation.py. It should support both European option puts and calls approximations. Tests should be in heston_approximation_test.py in the same folder.
-
Updated
Sep 12, 2020 - Rust
-
Updated
Jan 11, 2021 - C
-
Updated
Jan 4, 2021 - Python
-
Updated
Dec 31, 2020 - Clojure
-
Updated
Jan 10, 2021 - Python
-
Updated
Jan 10, 2021 - Nim
-
Updated
Jan 15, 2021 - HTML
-
Updated
Sep 8, 2018 - Shell
Speed up test suite
The standard accelerate test suite, used by all the backends, can be quite slow. Several of the tests are significantly slower than the others, for example segmented folds and scans, which I believe is because the reference implementations are very inefficient. Writing some more efficient reference implementations (e.g. using Data.Vector.Unboxed) should help speed things up.
-
Updated
Jan 19, 2021 - C++
-
Updated
Jan 19, 2021 - C++
Exchange a package that is marked as deprecated.
Seen in CI as warning in the apt install step with CUDA 10.1.243.
To Do:
- double check this works with CUDA 9.2 (current docker image; check minimal version constraint for Nvidia NGC) or...
- update Docker image to newer CUDA
Just an FYI whilst I was trawling through the ROCm GitHub page:
https://rocmdocs.amd.com/en/latest/Programming_Guides/Programming-Guides.html#
The problem is that the OpenCL types in https://github.com/triSYCL/triSYCL/blob/master/include/triSYCL/opencl_types.hpp are defined on the host according to the x86-64 Linux ABI which depends on the CPU & OS instead of using the description from https://www.khronos.org/registry/OpenCL/specs/2.2/html/OpenCL_C.html#built-in-scalar-data-types
Note that the system-wide cl_size_t has been removed
-
Updated
Sep 10, 2020 - Clojure
-
Updated
Dec 3, 2020 - C++
Recently #116 fixed a bug in OpAlgoBase to enable work groups which impeded the feautre to be used. A Python test was added to ensure that workgroups perform as expected:
https://github.com/EthicalML/vulkan-kompute/pull/116/files#diff-6e5055322f539aab58a242b92450beb2a97fc848f40ba9613ab5d720cff8a88bR116
This issue encompasses adding a C++ test that ensures the core feature of workgroups perform
-
Updated
Oct 9, 2018 - C++
-
Updated
Dec 22, 2020
-
Updated
Sep 26, 2020 - Clojure
- M: Mute (muting is not a node-wrangler feature, but I include it here because it's also node editor quality of life)
- Ctrl+Shift+LMB: View texture, material or volume node (create emission viewer if necessary)
- Ctrl+T: Create image node+attached mapping node
- Ctrl+Shift+T: Open file picker, user selects a bunch of textures, create disney material with textures attached to t
-
Updated
Dec 5, 2020 - C++
-
Updated
Dec 30, 2019 - Terra
-
Updated
Jan 14, 2021 - C++
-
Updated
Nov 27, 2020 - C++
Improve this page
Add a description, image, and links to the gpu-computing topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the gpu-computing topic, visit your repo's landing page and select "manage topics."
Problem: the approximate method can still be slow for many trees
catboost version: master
Operating System: ubuntu 18.04
CPU: i9
GPU: RTX2080
Would be good to be able to specify how many trees to use for shapley. The model.predict and prediction_type versions allow this. lgbm/xgb allow this.