AbstractPhil
/

geolip-svd-encoder-sweeps

Model card Files Files and versions

AbstractPhil commited on Apr 18

Commit

1c1e266

·

verified ·

1 Parent(s): 5933483

Update README.md

Files changed (1) hide show

README.md +2 -1

README.md CHANGED Viewed

@@ -7,7 +7,8 @@ license: apache-2.0
 from geolip_core import svd_transformer
-# S = S^N calculations of differentiation for utilization, default is 16 which is S^15
 # V = the the calculated embedding differentiation utilization size of the SVD, used for embedding and memory recall
 # D = the decomposition of the SVD, the size of the square matrix to decompose, this is the most computationally expensive part and should be kept as small as possible for large batch sizes, but it is also the most important for performance so it should not be too small. The sweet spot is usually around 16 or 32, but it can be smaller or larger depending on the use case and hardware.
 # Smaller D is specifically good for the triton kernels.

 from geolip_core import svd_transformer
+# S = S^N squarewise calculations of differentiation for utilization, default is 16 which is S^15
+  # Future versions will support rectangular after experimentation permits.
 # V = the the calculated embedding differentiation utilization size of the SVD, used for embedding and memory recall
 # D = the decomposition of the SVD, the size of the square matrix to decompose, this is the most computationally expensive part and should be kept as small as possible for large batch sizes, but it is also the most important for performance so it should not be too small. The sweet spot is usually around 16 or 32, but it can be smaller or larger depending on the use case and hardware.
 # Smaller D is specifically good for the triton kernels.