Tamas Bela Feher
Tamas Bela Feher
**Describe the bug** Approximate nearest neighbor models (`'ivfflat`', `'ivfpq'`) store their state is a` knnIndex` object. Currently there is no support to pickle models that were fitted using these algorithms....
**Describe the bug** #4743 describes a problem of saving model state. It is a general pattern that we see in other algorithms as well: we have some model state captured...
This PR moves the calibration function argument to the build function. In dynamic shape mode we need to provide profile information before we can calibrate the TensorRT engine. The `build()`...
This PR adds small changes to the benchmark scripts: - warmup steps - calculate mean, max and std time values.
**Describe the bug** When using a `Pipeline([ ('minmax', MinMaxScaler()), ('kmeans', Kmeans())])`, with a cudf `DataFrame` input, the results will be incorrect if the dataframe index is not sorted. **Steps/Code to...
**Is your feature request related to a problem? Please describe.** cuML provides pairwise distance metrics https://github.com/rapidsai/cuml/pull/2502 For large datasets GPU memory can becomes a limitation, and chunked pairwise distances would...
RAFT has a few parameters to control index building and search that does not have an equivalent public user parameter in FAISS. This issue is created to discuss whether we...
# Problem statement To get the best performance from RAFT ANN algorithms, it is recommended to use a pooling allocator. This is beneficial for the temporary allocations during the search...
Too long index file name would lead to a crash while calling the index serialization routines. Such long filenames can occur if we try to specialize many parameters for CAGRA...
Random sampling of training set for IVF methods was reverted in https://github.com/rapidsai/raft/pull/2144 due to the large memory usage of the subsample method. PR #2155 implements a new random sampling method....