Skip to content

Tessero\Laravel\TesseroManager

← All classes

AI assistant

Build this with Claude Code, Cursor or Copilot

Copy a Tessero-tuned prompt for Tessero\Laravel\TesseroManager, grounded in 15 real API signatures , into your IDE's AI. No chatbot, just exact context.

__construct
__construct(array $config = [])
backend
backend(): string

'ext' or 'ffi'

applyRuntimeSettings
applyRuntimeSettings(): void

Push threads / budget / tolerance into the backend (called at boot and per Octane request).

array
array(mixed $data, ?string $dtype = null): Tessero\Ext\NDArray|Tessero\NDArray
zeros
zeros(array|int $shape, string $dtype = 'float64'): Tessero\Ext\NDArray|Tessero\NDArray
fromBytes
fromBytes(string $bytes, string $dtype, array $shape): Tessero\Ext\NDArray|Tessero\NDArray
linprog
linprog(mixed $c, mixed $A_ub = null, mixed $b_ub = null, mixed $A_eq = null, mixed $b_eq = null, ?array $bounds = null): array

scipy.optimize.linprog. Returns x, fun, success, status, message, nit, ineqlin, eqlin (marginals).

milp
milp(mixed $c, array|bool $integrality = true, mixed $A_ub = null, mixed $b_ub = null, mixed $A_eq = null, mixed $b_eq = null, ?array $bounds = null): array

scipy.optimize.milp. Returns x, fun, success, status, message, nodes, best_bound.

solveMdp
solveMdp(mixed $P, mixed $R, float $gamma, string $method = 'policy_iteration'): array

Solve a discounted MDP. $P: (actions, states, states), $R: (states, actions). $method: 'policy_iteration' (exact) or 'value_iteration'.

normal
normal(int $seed, array|int $shape, float $mean = 0.0, float $sd = 1.0): Tessero\Ext\NDArray|Tessero\NDArray

NumPy-identical random numbers: numpy.random.default_rng($seed).normal(...).

random
random(int $seed, array|int $shape): Tessero\Ext\NDArray|Tessero\NDArray
memmap
memmap(string $filename, string $mode = 'r+', ?array $shape = null, string $dtype = 'float64', int $offset = 0): Tessero\Ext\NDArray|Tessero\NDArray

A memory-mapped array: the file's pages are the array's memory, so files larger than RAM can be processed. Modes as numpy.memmap: 'r', 'r+', 'w+', 'c'. Mapped bytes are not counted against memory_budget; info() reports them as memory_mapped. Both backends share libtessero's mapping code.

load
load(string $path, ?string $mmapMode = null): Tessero\Ext\NDArray|Tessero\NDArray

Read a .npy file (numpy.load); with $mmapMode ('r', 'r+', 'c') it is memory-mapped instead.

openMemmap
openMemmap(string $path, string $mode = 'r+', ?array $shape = null, string $dtype = 'float64', bool $fortranOrder = false): Tessero\Ext\NDArray|Tessero\NDArray

A memory-mapped .npy file (numpy.lib.format.open_memmap); 'w+' creates it.

info
info(): array
config
config(): array