The following options are supported.
Option name |
Type |
Default |
Description |
Constraints |
|---|---|---|---|---|
mixed precision |
string |
\(s=\) no |
Whether to use mixed precision iterative refinement, in which lower precision arithmetic is used before switching to the working precision for the final iterations. |
\(s=\) no, or yes. |
init |
string |
\(s=\) pca |
Initialization method for the embedding. |
\(s=\) pca, random, or supplied. |
low precision max_iter |
integer |
\(i=200\) |
If mixed precision iterative refinement is enabled, maximum number of iterations for the low precision phase. |
\(1 \le i\) |
check data |
string |
\(s=\) no |
Check input data for NaNs prior to performing computation. |
\(s=\) no, or yes. |
n_components |
integer |
\(i=2\) |
Number of embedding dimensions. |
\(1 \le i \le 3\) |
max_iter |
integer |
\(i=1000\) |
Maximum number of gradient descent iterations. |
\(1 \le i\) |
seed |
integer |
\(i=0\) |
Seed for random number generation; set to -1 for non-deterministic results. |
\(-1 \le i\) |
perplexity |
real |
\(r=30\) |
Target perplexity for conditional probabilities. |
\(1 \le r\) |
n_iter_without_progress |
integer |
\(i=300\) |
Stop if no progress is made for this many iterations. |
\(0 \le i\) |
early exaggeration |
real |
\(r=12\) |
Exaggeration factor for early iterations. |
\(1 \le r\) |
storage order |
string |
\(s=\) column-major |
Whether data is supplied and returned in row- or column-major order. |
\(s=\) c, column-major, f, fortran, or row-major. |
learning rate |
real |
\(r=-1\) |
Gradient descent learning rate. Use any non-positive value for auto: max(N / early_exaggeration / 4, 50). |
There are no constraints on \(r\). |
min_grad_norm |
real |
\(r=1e-07\) |
Stop if the gradient norm is below this threshold. |
\(0 \le r\) |
theta |
real |
\(r=0.5\) |
Barnes-Hut approximation parameter (0 for exact). |
\(0 \le r \le 1\) |
low precision min_grad_norm |
real |
\(r=0.0001\) |
If mixed precision iterative refinement is enabled, gradient norm convergence threshold for the low precision phase. |
\(0 \le r\) |