def test_upsampling_bilinear():
x = sym.Variable("x")
scale = 2
y = sym.upsampling(x,
scale=scale,
method="BILINEAR",
name="y",
layout="NCHW")
dtype = "float32"
dshape = (1, 4, 32, 32)
oshape = (1, 4, 32 * scale, 32 * scale)
shape_dict = {"x": dshape}
dtype_dict = {"x": dtype}
for target, ctx in ctx_list():
graph, lib, _ = nnvm.compiler.build(y, target, shape_dict, dtype_dict)
m = graph_runtime.create(graph, lib, ctx)
a_np = np.random.uniform(size=dshape).astype(dtype)
data = tvm.nd.array(a_np)
m.run(x=data)
out = m.get_output(0, tvm.nd.empty(oshape, dtype))
b_np = topi.testing.bilinear_resize_python(a_np,
(32 * scale, 32 * scale),
"NCHW",
align_corners=False)
tvm.testing.assert_allclose(out.asnumpy(), b_np, rtol=1e-5, atol=1e-5)
def _impl(cls, inputs, args, params):
width_scale = args['width_scale'] if 'width_scale' in args else 1
height_scale = args['height_scale'] if 'height_scale' in args else 1
assert width_scale == height_scale
return _sym.upsampling(
inputs[0], scale=int(width_scale), method="NEAREST_NEIGHBOR")
def get_sym(layout, kernel_layout, channels):
data = sym.Variable(name="data")
data = sym.conv2d(data=data, kernel_size=(3,3), channels=channels, padding=(1, 1),
layout=layout, kernel_layout=kernel_layout, use_bias=True)
data = sym.max_pool2d(data=data, pool_size=(2, 2), strides=(2, 2), layout=layout)
data = sym.upsampling(data=data, scale=2, layout=layout)
softmax_axis = 1
if layout == "NHWC":
softmax_axis = 3
data = sym.softmax(data=data, axis=softmax_axis)
return data
def get_sym(layout, kernel_layout, channels):
data = sym.Variable(name="data")
data = sym.conv2d(data=data, kernel_size=(3,3), channels=channels, padding=(1, 1),
layout=layout, kernel_layout=kernel_layout, use_bias=True)
data = sym.max_pool2d(data=data, pool_size=(2, 2), strides=(2, 2), layout=layout)
data = sym.upsampling(data=data, scale=2, layout=layout)
softmax_axis = 1
if layout == "NHWC":
softmax_axis = 3
data = sym.softmax(data=data, axis=softmax_axis)
return data
def test_upsampling_nearest_neighbor():
x = sym.Variable("x")
scale = 2
y = sym.upsampling(x, scale=scale, name="y")
dtype = "float32"
dshape = (1, 16, 32, 32)
oshape = (1, 16, 32*scale, 32*scale)
shape_dict = {"x": dshape}
for target, ctx in ctx_list():
graph, lib, _ = nnvm.compiler.build(y, target, shape_dict)
m = graph_runtime.create(graph, lib, ctx)
a_np = np.random.uniform(size=dshape).astype(dtype)
data = tvm.nd.array(a_np)
m.run(x=data)
out = m.get_output(0, tvm.nd.empty(oshape, dtype))
b_np = topi.testing.upsampling_python(a_np, scale, "NCHW")
np.testing.assert_allclose(out.asnumpy(), b_np, rtol=1e-5)
def test_upsampling_nearest_neighbor():
x = sym.Variable("x")
scale = 2
y = sym.upsampling(x, scale=scale, name="y")
dtype = "float32"
dshape = (1, 16, 32, 32)
oshape = (1, 16, 32*scale, 32*scale)
shape_dict = {"x": dshape}
for target, ctx in ctx_list():
graph, lib, _ = nnvm.compiler.build(y, target, shape_dict)
m = graph_runtime.create(graph, lib, ctx)
a_np = np.random.uniform(size=dshape).astype(dtype)
data = tvm.nd.array(a_np)
m.run(x=data)
out = m.get_output(0, tvm.nd.empty(oshape, dtype))
b_np = topi.testing.upsampling_python(a_np, scale, "NCHW")
tvm.testing.assert_allclose(out.asnumpy(), b_np, rtol=1e-5)
