09_onnx_build.py

Go to the documentation of this file.

import numpy as np

import onnx
from onnx import helper, shape_inference, optimizer
from onnx import numpy_helper
from onnx import AttributeProto, TensorProto, GraphProto

# Create graph input X
X = helper.make_tensor_value_info('X', TensorProto.FLOAT, [1, 1, 8, 8])

W_info = helper.make_tensor_value_info('W', TensorProto.FLOAT, [2, 1, 3, 3])
W = np.array([[[[-1,-1,-1],[-1,8,-1],[-1,-1,-1]]],[[[0,-1,0],[-1,5,-1],[0,-1,0]]]]).astype(np.float32)
W = numpy_helper.from_array(W, 'W')


# Create graph output Y
Y = helper.make_tensor_value_info('Y', TensorProto.FLOAT, [1, 2, 8, 8])

conv1 = helper.make_node(
        'Conv',         # name
        ['X', 'W'],     # inputs
        ['Y'],          # outputs
        
        kernel_shape=[3, 3],
        strides=[1, 1],
        pads=[1, 1, 1, 1],
        )


graph_def = helper.make_graph(
        nodes=[conv1],     # graph nodes
        name= 'conv2d_a_model',   # graph name
        inputs =      [X, W_info],  # graph inputs
        outputs =     [Y],          # graph outputs
        initializer = [W],
        )

model_def = helper.make_model(graph_def, producer_name='benchmarks')

onnx.checker.check_model(model_def)
model_def = shape_inference.infer_shapes(model_def)
onnx.checker.check_model(model_def)
model_def = optimizer.optimize(model_def)
onnx.checker.check_model(model_def)

onnx.save_model(model_def, '09_conv2d_a.onnx')