deploy_model_on_rasp.py 8.06 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements.  See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership.  The ASF licenses this file
# to you under the Apache License, Version 2.0 (the
# "License"); you may not use this file except in compliance
# with the License.  You may obtain a copy of the License at
#
#   http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
# KIND, either express or implied.  See the License for the
# specific language governing permissions and limitations
# under the License.
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
"""
.. _tutorial-deploy-model-on-rasp:

Deploy the Pretrained Model on Raspberry Pi
===========================================
**Author**: `Ziheng Jiang <https://ziheng.org/>`_, \
            `Hiroyuki Makino <https://makihiro.github.io/>`_

This is an example of using Relay to compile a ResNet model and deploy
it on Raspberry Pi.
"""

import tvm
import tvm.relay as relay
from tvm import rpc
from tvm.contrib import util, graph_runtime as runtime
33
from tvm.contrib.download import download_testdata
34 35 36 37 38 39 40

######################################################################
# .. _build-tvm-runtime-on-device:
#
# Build TVM Runtime on Device
# ---------------------------
#
41
# The first step is to build the TVM runtime on the remote device.
42 43 44 45 46 47
#
# .. note::
#
#   All instructions in both this section and next section should be
#   executed on the target device, e.g. Raspberry Pi. And we assume it
#   has Linux running.
48
#
49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64
# Since we do compilation on local machine, the remote device is only used
# for running the generated code. We only need to build tvm runtime on
# the remote device.
#
# .. code-block:: bash
#
#   git clone --recursive https://github.com/dmlc/tvm
#   cd tvm
#   mkdir build
#   cp cmake/config.cmake build
#   cd build
#   cmake ..
#   make runtime -j4
#
# After building runtime successfully, we need to set environment varibles
# in :code:`~/.bashrc` file. We can edit :code:`~/.bashrc`
65
# using :code:`vi ~/.bashrc` and add the line below (Assuming your TVM
66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95
# directory is in :code:`~/tvm`):
#
# .. code-block:: bash
#
#   export PYTHONPATH=$PYTHONPATH:~/tvm/python
#
# To update the environment variables, execute :code:`source ~/.bashrc`.

######################################################################
# Set Up RPC Server on Device
# ---------------------------
# To start an RPC server, run the following command on your remote device
# (Which is Raspberry Pi in our example).
#
#   .. code-block:: bash
#
#     python -m tvm.exec.rpc_server --host 0.0.0.0 --port=9090
#
# If you see the line below, it means the RPC server started
# successfully on your device.
#
#    .. code-block:: bash
#
#      INFO:root:RPCServer: bind to 0.0.0.0:9090
#

######################################################################
# Prepare the Pre-trained Model
# -----------------------------
# Back to the host machine, which should have a full TVM installed (with LLVM).
96
#
97 98 99 100 101 102 103 104 105 106 107 108 109 110
# We will use pre-trained model from
# `MXNet Gluon model zoo <https://mxnet.incubator.apache.org/api/python/gluon/model_zoo.html>`_.
# You can found more details about this part at tutorial :ref:`tutorial-from-mxnet`.

from mxnet.gluon.model_zoo.vision import get_model
from PIL import Image
import numpy as np

# one line to get the model
block = get_model('resnet18_v1', pretrained=True)

######################################################################
# In order to test our model, here we download an image of cat and
# transform its format.
111
img_url = 'https://github.com/dmlc/mxnet.js/blob/master/data/cat.png?raw=true'
112
img_name = 'cat.png'
113 114
img_path = download_testdata(img_url, img_name, module='data')
image = Image.open(img_path).resize((224, 224))
115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131

def transform_image(image):
    image = np.array(image) - np.array([123., 117., 104.])
    image /= np.array([58.395, 57.12, 57.375])
    image = image.transpose((2, 0, 1))
    image = image[np.newaxis, :]
    return image

x = transform_image(image)

######################################################################
# synset is used to transform the label from number of ImageNet class to
# the word human can understand.
synset_url = ''.join(['https://gist.githubusercontent.com/zhreshold/',
                      '4d0b62f3d01426887599d4f7ede23ee5/raw/',
                      '596b27d23537e5a1b5751d2b0481ef172f58b539/',
                      'imagenet1000_clsid_to_human.txt'])
132 133 134
synset_name = 'imagenet1000_clsid_to_human.txt'
synset_path = download_testdata(synset_url, synset_name, module='data')
with open(synset_path) as f:
135 136 137 138 139 140 141 142
    synset = eval(f.read())

######################################################################
# Now we would like to port the Gluon model to a portable computational graph.
# It's as easy as several lines.

# We support MXNet static graph(symbol) and HybridBlock in mxnet.gluon
shape_dict = {'data': x.shape}
143
mod, params = relay.frontend.from_mxnet(block, shape_dict)
144
# we want a probability so add a softmax operator
145
func = mod[mod.entry_func]
146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225
func = relay.Function(func.params, relay.nn.softmax(func.body), None, func.type_params, func.attrs)

######################################################################
# Here are some basic data workload configurations.
batch_size = 1
num_classes = 1000
image_shape = (3, 224, 224)
data_shape = (batch_size,) + image_shape

######################################################################
# Compile The Graph
# -----------------
# To compile the graph, we call the :any:`relay.build` function
# with the graph configuration and parameters. However, You cannot to
# deploy a x86 program on a device with ARM instruction set. It means
# Relay also needs to know the compilation option of target device,
# apart from arguments :code:`net` and :code:`params` to specify the
# deep learning workload. Actually, the option matters, different option
# will lead to very different performance.

######################################################################
# If we run the example on our x86 server for demonstration, we can simply
# set it as :code:`llvm`. If running it on the Raspberry Pi, we need to
# specify its instruction set. Set :code:`local_demo` to False if you want
# to run this tutorial with a real device.

local_demo = True

if local_demo:
    target = tvm.target.create('llvm')
else:
    target = tvm.target.arm_cpu('rasp3b')
    # The above line is a simple form of
    # target = tvm.target.create('llvm -device=arm_cpu -model=bcm2837 -target=armv7l-linux-gnueabihf -mattr=+neon')

with relay.build_config(opt_level=3):
    graph, lib, params = relay.build(func, target, params=params)

# After `relay.build`, you will get three return values: graph,
# library and the new parameter, since we do some optimization that will
# change the parameters but keep the result of model as the same.

# Save the library at local temporary directory.
tmp = util.tempdir()
lib_fname = tmp.relpath('net.tar')
lib.export_library(lib_fname)

######################################################################
# Deploy the Model Remotely by RPC
# --------------------------------
# With RPC, you can deploy the model remotely from your host machine
# to the remote device.

# obtain an RPC session from remote device.
if local_demo:
    remote = rpc.LocalSession()
else:
    # The following is my environment, change this to the IP address of your target device
    host = '10.77.1.162'
    port = 9090
    remote = rpc.connect(host, port)

# upload the library to remote device and load it
remote.upload(lib_fname)
rlib = remote.load_module('net.tar')

# create the remote runtime module
ctx = remote.cpu(0)
module = runtime.create(graph, rlib, ctx)
# set parameter (upload params to the remote device. This may take a while)
module.set_input(**params)
# set input data
module.set_input('data', tvm.nd.array(x.astype('float32')))
# run
module.run()
# get output
out = module.get_output(0)
# get top1 result
top1 = np.argmax(out.asnumpy())
print('TVM prediction top-1: {}'.format(synset[top1]))