from tvm import relay
from tvm.relay import testing
import numpy as np
from infrastructure import get_ref_result
batch_size = 1
num_class = 1000
image_shape = (3, 224, 224)
data_shape = (batch_size,) + image_shape
out_shape = (batch_size, num_class)
dtype="float32"
mod, params = relay.testing.mobilenet.get_workload(
batch_size=batch_size, num_classes=num_class, image_shape=image_shape,
layout='NCHW'
)
#print(mod.astext(show_meta_data=False))
data = np.random.uniform(size=data_shape).astype(dtype)
print(data)
ref_out = get_ref_result(data, mod, params, out_shape, dtype)
print(ref_out)
get_ref_result looks like:
def get_ref_result(data, mod, params, out_shape, dtype):
target = "llvm"
with tvm.transform.PassContext(opt_level=3,
disabled_pass=["AlterOpLayout"]):
lib = relay.build(mod, target, params=params)
cpu_mod = graph_runtime.GraphModule(lib["default"](tvm.cpu()))
cpu_mod.set_input("data", data)
cpu_mod.run()
cpu_out = cpu_mod.get_output(0, tvm.nd.empty(out_shape, dtype))
return cpu_out
output:
================================================================
[[[[0.48127168 0.2018001 0.71724653 ... 0.0441279 0.57116777
0.1731153 ]
[0.40417442 0.3016946 0.74636394 ... 0.3417648 0.718218
0.28890228]
[0.7683302 0.17131594 0.9016031 ... 0.5153679 0.74072677
0.03374053]
...
[0.6947896 0.6551721 0.85114497 ... 0.35421443 0.20508686
0.6471268 ]
[0.09923462 0.61146086 0.08773897 ... 0.53768474 0.31748652
0.64678025]
[0.31008628 0.56266195 0.83621436 ... 0.9968801 0.4973068
0.09383171]]
[[0.73113763 0.17166294 0.5789204 ... 0.03240918 0.0247721
0.89045954]
[0.46058905 0.3739123 0.56078994 ... 0.38859197 0.36561185
0.7287658 ]
[0.8079502 0.39894798 0.6348208 ... 0.56089103 0.58005774
0.52373666]
...
[0.4517257 0.8520253 0.40640992 ... 0.1651029 0.22171977
0.35451823]
[0.9394899 0.7759206 0.5117806 ... 0.99209446 0.24618751
0.57113916]
[0.6102327 0.08231816 0.7101693 ... 0.77034265 0.9671634
0.5752965 ]]
[[0.3101213 0.192366 0.22534423 ... 0.828487 0.59424293
0.21207647]
[0.8794648 0.09954574 0.30758655 ... 0.051931 0.03809953
0.3480195 ]
[0.81616604 0.92345166 0.36221072 ... 0.93277586 0.79536366
0.42082992]
...
[0.621181 0.4233806 0.83933717 ... 0.44883785 0.4910011
0.3370444 ]
[0.9489613 0.7982109 0.709624 ... 0.6371652 0.5758706
0.6982647 ]
[0.36476108 0.2929088 0.49834147 ... 0.87037426 0.40084326
0.3614452 ]]]]
Cannot find config for target=llvm -keys=cpu, workload=('dense_nopack.x86',
('TENSOR', (1, 1024), 'float32'), ('TENSOR', (1000, 1024), 'float32'), None,
'float32'). A fallback configuration is used, which may bring great performance
regression.
Cannot find config for target=llvm -keys=cpu, workload=('conv2d_NCHWc.x86',
('TENSOR', (1, 1024, 7, 7), 'float32'), ('TENSOR', (1024, 1024, 1, 1),
'float32'), (1, 1), (0, 0, 0, 0), (1, 1), 'NCHW', 'NCHW', 'float32'). A
fallback configuration is used, which may bring great performance regression.
Cannot find config for target=llvm -keys=cpu,
workload=('depthwise_conv2d_NCHWc.x86', ('TENSOR', (1, 1024, 7, 7), 'float32'),
('TENSOR', (1024, 1, 3, 3), 'float32'), (1, 1), (1, 1, 1, 1), (1, 1), 'NCHW',
'NCHW', 'float32'). A fallback configuration is used, which may bring great
performance regression.
Cannot find config for target=llvm -keys=cpu, workload=('conv2d_NCHWc.x86',
('TENSOR', (1, 512, 7, 7), 'float32'), ('TENSOR', (1024, 512, 1, 1),
'float32'), (1, 1), (0, 0, 0, 0), (1, 1), 'NCHW', 'NCHW', 'float32'). A
fallback configuration is used, which may bring great performance regression.
Cannot find config for target=llvm -keys=cpu,
workload=('depthwise_conv2d_NCHWc.x86', ('TENSOR', (1, 512, 14, 14),
'float32'), ('TENSOR', (512, 1, 3, 3), 'float32'), (2, 2), (1, 1, 1, 1), (1,
1), 'NCHW', 'NCHW', 'float32'). A fallback configuration is used, which may
bring great performance regression.
Cannot find config for target=llvm -keys=cpu, workload=('conv2d_NCHWc.x86',
('TENSOR', (1, 512, 14, 14), 'float32'), ('TENSOR', (512, 512, 1, 1),
'float32'), (1, 1), (0, 0, 0, 0), (1, 1), 'NCHW', 'NCHW', 'float32'). A
fallback configuration is used, which may bring great performance regression.
Cannot find config for target=llvm -keys=cpu,
workload=('depthwise_conv2d_NCHWc.x86', ('TENSOR', (1, 512, 14, 14),
'float32'), ('TENSOR', (512, 1, 3, 3), 'float32'), (1, 1), (1, 1, 1, 1), (1,
1), 'NCHW', 'NCHW', 'float32'). A fallback configuration is used, which may
bring great performance regression.
Cannot find config for target=llvm -keys=cpu, workload=('conv2d_NCHWc.x86',
('TENSOR', (1, 256, 14, 14), 'float32'), ('TENSOR', (512, 256, 1, 1),
'float32'), (1, 1), (0, 0, 0, 0), (1, 1), 'NCHW', 'NCHW', 'float32'). A
fallback configuration is used, which may bring great performance regression.
Cannot find config for target=llvm -keys=cpu,
workload=('depthwise_conv2d_NCHWc.x86', ('TENSOR', (1, 256, 28, 28),
'float32'), ('TENSOR', (256, 1, 3, 3), 'float32'), (2, 2), (1, 1, 1, 1), (1,
1), 'NCHW', 'NCHW', 'float32'). A fallback configuration is used, which may
bring great performance regression.
Cannot find config for target=llvm -keys=cpu, workload=('conv2d_NCHWc.x86',
('TENSOR', (1, 256, 28, 28), 'float32'), ('TENSOR', (256, 256, 1, 1),
'float32'), (1, 1), (0, 0, 0, 0), (1, 1), 'NCHW', 'NCHW', 'float32'). A
fallback configuration is used, which may bring great performance regression.
Cannot find config for target=llvm -keys=cpu,
workload=('depthwise_conv2d_NCHWc.x86', ('TENSOR', (1, 256, 28, 28),
'float32'), ('TENSOR', (256, 1, 3, 3), 'float32'), (1, 1), (1, 1, 1, 1), (1,
1), 'NCHW', 'NCHW', 'float32'). A fallback configuration is used, which may
bring great performance regression.
Cannot find config for target=llvm -keys=cpu, workload=('conv2d_NCHWc.x86',
('TENSOR', (1, 128, 28, 28), 'float32'), ('TENSOR', (256, 128, 1, 1),
'float32'), (1, 1), (0, 0, 0, 0), (1, 1), 'NCHW', 'NCHW', 'float32'). A
fallback configuration is used, which may bring great performance regression.
Cannot find config for target=llvm -keys=cpu,
workload=('depthwise_conv2d_NCHWc.x86', ('TENSOR', (1, 128, 56, 56),
'float32'), ('TENSOR', (128, 1, 3, 3), 'float32'), (2, 2), (1, 1, 1, 1), (1,
1), 'NCHW', 'NCHW', 'float32'). A fallback configuration is used, which may
bring great performance regression.
Cannot find config for target=llvm -keys=cpu, workload=('conv2d_NCHWc.x86',
('TENSOR', (1, 128, 56, 56), 'float32'), ('TENSOR', (128, 128, 1, 1),
'float32'), (1, 1), (0, 0, 0, 0), (1, 1), 'NCHW', 'NCHW', 'float32'). A
fallback configuration is used, which may bring great performance regression.
Cannot find config for target=llvm -keys=cpu,
workload=('depthwise_conv2d_NCHWc.x86', ('TENSOR', (1, 128, 56, 56),
'float32'), ('TENSOR', (128, 1, 3, 3), 'float32'), (1, 1), (1, 1, 1, 1), (1,
1), 'NCHW', 'NCHW', 'float32'). A fallback configuration is used, which may
bring great performance regression.
Cannot find config for target=llvm -keys=cpu, workload=('conv2d_NCHWc.x86',
('TENSOR', (1, 64, 56, 56), 'float32'), ('TENSOR', (128, 64, 1, 1), 'float32'),
(1, 1), (0, 0, 0, 0), (1, 1), 'NCHW', 'NCHW', 'float32'). A fallback
configuration is used, which may bring great performance regression.
Cannot find config for target=llvm -keys=cpu,
workload=('depthwise_conv2d_NCHWc.x86', ('TENSOR', (1, 64, 112, 112),
'float32'), ('TENSOR', (64, 1, 3, 3), 'float32'), (2, 2), (1, 1, 1, 1), (1, 1),
'NCHW', 'NCHW', 'float32'). A fallback configuration is used, which may bring
great performance regression.
Cannot find config for target=llvm -keys=cpu, workload=('conv2d_NCHWc.x86',
('TENSOR', (1, 32, 112, 112), 'float32'), ('TENSOR', (64, 32, 1, 1),
'float32'), (1, 1), (0, 0, 0, 0), (1, 1), 'NCHW', 'NCHW', 'float32'). A
fallback configuration is used, which may bring great performance regression.
Cannot find config for target=llvm -keys=cpu,
workload=('depthwise_conv2d_NCHWc.x86', ('TENSOR', (1, 32, 112, 112),
'float32'), ('TENSOR', (32, 1, 3, 3), 'float32'), (1, 1), (1, 1, 1, 1), (1, 1),
'NCHW', 'NCHW', 'float32'). A fallback configuration is used, which may bring
great performance regression.
Cannot find config for target=llvm -keys=cpu, workload=('conv2d_NCHWc.x86',
('TENSOR', (1, 3, 224, 224), 'float32'), ('TENSOR', (32, 3, 3, 3), 'float32'),
(2, 2), (1, 1, 1, 1), (1, 1), 'NCHW', 'NCHW', 'float32'). A fallback
configuration is used, which may bring great performance regression.
[[0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
0.001 0.001 0.001 0.001]]
---
[Visit
Topic](https://discuss.tvm.apache.org/t/why-the-mobilenet-workload-result-is-all-0-001/8420/1)
to respond.
You are receiving this because you enabled mailing list mode.
To unsubscribe from these emails, [click
here](https://discuss.tvm.apache.org/email/unsubscribe/b47f7234fdfe9346d9242f3f911cb6bf93d64344d7625b75f239eb035e325a54).
