Skip to content

Reference for ultralytics/nn/backends/mnn.py

Improvements

This page is sourced from https://github.com/ultralytics/ultralytics/blob/main/ultralytics/nn/backends/mnn.py. Have an improvement or example to add? Open a Pull Request — thank you! 🙏


Summary

class ultralytics.nn.backends.mnn.MNNBackend

MNNBackend()

Bases: BaseBackend

MNN (Mobile Neural Network) inference backend.

Loads and runs inference with MNN models (.mnn files) using the Alibaba MNN framework. Optimized for mobile and edge deployment with configurable thread count and precision.

Methods

NameDescription
forwardRun inference using the MNN runtime.
load_modelLoad an Alibaba MNN model from a .mnn file.
Source code in ultralytics/nn/backends/mnn.pyView on GitHub
class MNNBackend(BaseBackend):


method ultralytics.nn.backends.mnn.MNNBackend.forward

def forward(self, im: torch.Tensor) -> list

Run inference using the MNN runtime.

Args

NameTypeDescriptionDefault
imtorch.TensorInput image tensor in BCHW format, normalized to [0, 1].required

Returns

TypeDescription
listModel predictions as a list of numpy arrays.
Source code in ultralytics/nn/backends/mnn.pyView on GitHub
def forward(self, im: torch.Tensor) -> list:
    """Run inference using the MNN runtime.

    Args:
        im (torch.Tensor): Input image tensor in BCHW format, normalized to [0, 1].

    Returns:
        (list): Model predictions as a list of numpy arrays.
    """
    input_var = self.expr.const(im.data_ptr(), im.shape)
    output_var = self.net.onForward([input_var])
    # NOTE: need this copy(), or it'd get incorrect results on ARM devices
    return [x.read().copy() for x in output_var]


method ultralytics.nn.backends.mnn.MNNBackend.load_model

def load_model(self, weight: str | Path) -> None

Load an Alibaba MNN model from a .mnn file.

Args

NameTypeDescriptionDefault
weightstr | PathPath to the .mnn model file.required
Source code in ultralytics/nn/backends/mnn.pyView on GitHub
def load_model(self, weight: str | Path) -> None:
    """Load an Alibaba MNN model from a .mnn file.

    Args:
        weight (str | Path): Path to the .mnn model file.
    """
    LOGGER.info(f"Loading {weight} for MNN inference...")
    check_requirements("MNN")
    import MNN

    config = {"precision": "low", "backend": "CPU", "numThread": (os.cpu_count() + 1) // 2}
    rt = MNN.nn.create_runtime_manager((config,))
    self.net = MNN.nn.load_module_from_file(weight, [], [], runtime_manager=rt, rearrange=True)
    self.expr = MNN.expr

    # Load metadata from bizCode
    info = self.net.get_info()
    if "bizCode" in info:
        try:
            self.apply_metadata(json.loads(info["bizCode"]))
        except json.JSONDecodeError:
            pass





📅 Created 0 days ago ✏️ Updated 0 days ago
Laughing-q