onnx-array-api: (Numpy) Array API for ONNX

onnx-array-api implements a numpy API for ONNX. It gives the user the ability to convert functions written following the numpy API to convert that function into ONNX as well as to execute it.

• Tutorial
  • Numpy API for ONNX
  • Benchmarks
• API
  • npx.npx_functions
  • npx.npx_var
  • npx, jit and eager mode
  • npx.npx_types
  • npx.npx_numpy_tensors
  • ort
  • plotting
  • tools
• Example gallery

Sources available on github/onnx-array-api, see also code coverage.

<<<

import numpy as np  # A
from onnx_array_api.npx import absolute, jit_onnx
from onnx_array_api.plotting.dot_plot import to_dot


def l1_loss(x, y):
    return absolute(x - y).sum()


def l2_loss(x, y):
    return ((x - y) ** 2).sum()


def myloss(x, y):
    return l1_loss(x[:, 0], y[:, 0]) + l2_loss(x[:, 1], y[:, 1])


jitted_myloss = jit_onnx(myloss)

x = np.array([[0.1, 0.2], [0.3, 0.4]], dtype=np.float32)
y = np.array([[0.11, 0.22], [0.33, 0.44]], dtype=np.float32)
res = jitted_myloss(x, y)
print(res)

>>>

    [0.042]