Anupam
Anupam
Hi, I am using the sample script in this repository `resnet50_ddp_profiler.py` from https://github.com/pytorch/kineto/blob/main/tb_plugin/examples/resnet50_ddp_profiler.py Using ``` Python3.8 torch=2.0.1 torch-tb-profiler=0.4.3 # built from source ``` In tensorboard in the overview view the...
In the middle of adding the PCA initialization for TSNE, we realized that there is no support yet for sparse inputs for PCA.
Closes https://github.com/rapidsai/cuml/issues/3458 Add PCA embedding initialization to C++ layer and expose it in Python API. ```python from cuml.manifold import TSNE tsne = TSNE( ... init="pca" # ("random" or "pca") )...
Closes https://github.com/rapidsai/cuml/issues/5129 Related to https://github.com/rapidsai/raft/pull/2359 1. Enables dice metric for UMAP 2. Enables dice metric for `cuml.metrics.pairwise_distances`
### Description The eigenvalues when using cupy eigsh are NaN and not the same as scipy eigsh. By inspecting the output in the _lanczos_fast algorithm that cupy uses, the output...
Documentation for future work: * The new lanczos solver in raft https://github.com/rapidsai/raft/pull/2416 can be integrated into the UMAP spectral initialization algorithm. * After the UMAP_::get_graph step the resulting COO matrix...
## Lanczos Solver for Sparse Eigen Decomposition We propose a new lanczos solver in raft that fixes the issues present in the previous solver `raft::sparse::solver::detail::computeSmallestEigenvectors`. Specifically we address the following...
Documenting for future work: Given a sparse matrix `A` convert it to a sparse laplacian matrix `L` where `D` is degree matrix. `L = D - A` Ideas on how...
Resolves #6912 # Usage ```python from sklearn import datasets from cuml.manifold import SpectralEmbedding import cupy as cp # (1500, 3) -> (1500, 2) n_samples = 1500 S_points, S_color = datasets.make_s_curve(n_samples,...