mirror of
https://github.com/luxfi/torus-ml.git
synced 2026-10-30 15:15:23 +00:00
zeekay
e037e8f7ea
The distribution called itself torus-ml while shipping a package directory named concrete, so the name a user installs and the name they import disagreed — and most of the tree already said torus.ml, which could not resolve against src/concrete. Moved src/concrete to src/torus and pointed the include at it. The dependency was Zama's concrete-python from their index. It is torus-fhe from ours, which is what luxfhe/torus already ships: that frontend declares torus-fhe and ships a torus/fhe module, so every 'from torus.fhe' in this tree was already naming our compiler and only the dependency lagged. Two distributions share one namespace, which is the shape the upstream this forked from used and the reason the split works: torus-fhe gives torus.fhe, torus-ml gives torus.ml. Swept every other spelling — identifiers, enum members, workflow variables, filenames, notebooks. LICENSE and NOTICE keep the upstream attribution a fork owes; the figures carry the old name in JPEG metadata a text sweep cannot reach.
2.8 KiB
2.8 KiB
module torus.ml.quantization.linear_op_glwe_backend
GLWE backend for some supported layers.
function has_glwe_backend
has_glwe_backend()
Check if the GLWE backend is installed.
Returns:
bool: True if the GLWE backend is installed, False otherwise.
class GLWELinearLayerExecutor
GLWE execution helper for pure linear layers.
method __init__
__init__(
private_key=None,
compression_key=None,
use_dynamic_quantization: bool = True
)
Initialize the GLWE linear layer executor.
Args:
private_key: Private key for encryptioncompression_key: Compression key for encryptionuse_dynamic_quantization: Whether to use dynamic quantization
method forward
forward(x: Tensor, q_module: QuantizedModule, fhe: HybridFHEMode) → Tensor
Perform the inference of this linear layer.
Args:
x(torch.Tensor): Input or previous layer activations.q_module(QuantizedModule): Module containing the quantized layer.fhe(HybridFHEMode): Execution mode, can be 'DISABLE' or 'EXECUTE'.
Returns:
torch.Tensor: The result of applying the linear layer.
method keygen
keygen()
Generate private and compression key.