Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Tvmc python tutorial #9633

Merged
merged 32 commits into from
Jan 19, 2022
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
32 commits
Select commit Hold shift + click to select a range
4cc90ac
finish rpc and shape_dict in tut
Sep 3, 2021
60d2131
added more to rpc
Sep 13, 2021
0ea76cf
tutorial edits
Sep 15, 2021
ce6e8a6
added tutorial to docs in howto
CircleSpin Dec 2, 2021
b3ff2c9
accidentally had two copies of tutorial
CircleSpin Dec 2, 2021
00a0fda
Update gallery/how_to/use_tvms_python_api/tvmc_python.py
CircleSpin Dec 2, 2021
d83b579
Update gallery/how_to/use_tvms_python_api/tvmc_python.py
CircleSpin Dec 2, 2021
9bf946f
Update gallery/how_to/use_tvms_python_api/tvmc_python.py
CircleSpin Dec 2, 2021
71e5121
Update gallery/how_to/use_tvms_python_api/tvmc_python.py
CircleSpin Dec 2, 2021
6c4642c
Apply suggestions from code review
CircleSpin Dec 2, 2021
871da11
Update gallery/how_to/use_tvms_python_api/tvmc_python.py
CircleSpin Dec 2, 2021
0eaf858
Update gallery/how_to/use_tvms_python_api/tvmc_python.py
CircleSpin Dec 2, 2021
26135ac
added Leandro's suggestions
CircleSpin Dec 3, 2021
e621327
added example model at top
CircleSpin Dec 3, 2021
a92f9ed
added example model, blacked it
CircleSpin Dec 3, 2021
3123280
trying to get docs to build
CircleSpin Dec 3, 2021
0df1e56
underline too short for title
CircleSpin Dec 3, 2021
f14f400
forgot Jetson info, added Chris H comments
CircleSpin Dec 14, 2021
9f9d404
reformatting text
CircleSpin Dec 14, 2021
8ba2d61
black
CircleSpin Dec 15, 2021
4985e15
hitting code block issue, trying to debug
CircleSpin Dec 15, 2021
aee6db3
added spaces after the python codeblock
CircleSpin Dec 16, 2021
74b3826
black
CircleSpin Dec 16, 2021
45ac07e
changing formatting
CircleSpin Dec 18, 2021
0285d84
touching up more edits'
CircleSpin Dec 18, 2021
7fcc2c8
more touchups
CircleSpin Dec 18, 2021
1786f4a
changed location of file to tutorial section
CircleSpin Dec 18, 2021
33c6ea0
changing doc location
CircleSpin Dec 18, 2021
68ab249
broke the order of the docs somehow
CircleSpin Dec 18, 2021
cbf6cdc
fixed it yayy
CircleSpin Dec 18, 2021
846a6e1
added additional indentation
CircleSpin Jan 19, 2022
ac778a4
black'd
CircleSpin Jan 19, 2022
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions docs/conf.py
Original file line number Diff line number Diff line change
Expand Up @@ -255,6 +255,7 @@ def git_describe_version(original_version):
"introduction.py",
"install.py",
"tvmc_command_line_driver.py",
"tvmc_python.py",
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

+1 to this, thanks!

"autotvm_relay_x86.py",
"tensor_expr_get_started.py",
"autotvm_matmul_x86.py",
Expand Down
4 changes: 4 additions & 0 deletions docs/legacy_redirect.py
Original file line number Diff line number Diff line change
Expand Up @@ -242,6 +242,10 @@
"tutorials/get_started/tvmc_command_line_driver.html",
"../../tutorial/tvmc_command_line_driver.html",
],
[
"tutorials/get_started/tvmc_python.html",
"../../tutorial/tvmc_python.html",
],
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Since this a new page, a redirect isn't necessary.

]

redirect_template = """
Expand Down
292 changes: 292 additions & 0 deletions gallery/tutorial/tvmc_python.py
Original file line number Diff line number Diff line change
@@ -0,0 +1,292 @@
# 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.
"""
Getting Starting using TVMC Python: a high-level API for TVM
=============================================================
**Author**:
`Jocelyn Shiue <https://github.com/CircleSpin>`_

Hi! Here we explain the scripting tool designed for the complete TVM beginner. 🙂

Before we get started let's get an example model if you don't already have one.
Follow the steps to download a resnet model via the terminal:

.. code-block:: python

mkdir myscripts
cd myscripts
wget https://github.com/onnx/models/raw/master/vision/classification/resnet/model/resnet50-v2-7.onnx
mv resnet50-v2-7.onnx my_model.onnx
touch tvmcpythonintro.py

Let's start editing the python file in your favorite text editor.
"""

################################################################################
# Step 0: Imports
# ~~~~~~~~~~~~~~~
#
# .. code-block:: python
#
# from tvm.driver import tvmc
#
#

################################################################################
# Step 1: Load a model
# ~~~~~~~~~~~~~~~~~~~~
#
# Let's import our model into tvmc. This step converts a machine learning model from
# a supported framework into TVM's high level graph representation language called Relay.
# This is to have a unified starting point for all models in tvm. The frameworks we currently
# support are: Keras, ONNX, Tensorflow, TFLite, and PyTorch.
#
# .. code-block:: python
#
# model = tvmc.load('my_model.onnx') #Step 1: Load
#
# If you'd like to see the Relay, you can run:
# ``model.summary()``
#
# All frameworks support overwriting the input shapes with a shape_dict argument.
# For most frameworks this is optional, but for Pytorch this is necessary as
# TVM cannot automatically search for it.
#
# .. code-block:: python
#
# #model = tvmc.load(my_model, shape_dict={'input1' : [1, 2, 3, 4], 'input2' : [1, 2, 3, 4]}) #Step 1: Load + shape_dict
#
# A suggested way to see the model's input/shape_dict is via `netron <https://netron.app/>`_. After opening the model,
# click the first node to see the name(s) and shape(s) in the inputs section.


################################################################################
# Step 2: Compile
# ~~~~~~~~~~~~~~~
#
# Now that our model is in Relay, our next step is to compile it to a desired
# hardware to run on. We refer to this hardware as a target. This compilation process
# translates the model from Relay into a lower-level language that the
# target machine can understand.
#
# In order to compile a model a tvm.target string is required.
# To learn more about tvm.targets and their options look at the `documentation <https://tvm.apache.org/docs/api/python/target.html>`_.
# Some examples include:
#
# 1. cuda (Nvidia GPU)
# 2. llvm (CPU)
# 3. llvm -mcpu=cascadelake (Intel CPU)
#
# .. code-block:: python
#
# package = tvmc.compile(model, target="llvm") #Step 2: Compile
#
#
# The compilation step returns a package.
#

################################################################################
# Step 3: Run
# ~~~~~~~~~~~
#
# The compiled package can now be run on the hardware target. The device
# input options are: CPU, Cuda, CL, Metal, and Vulkan.
#
# .. code-block:: python
#
# result = tvmc.run(package, device="cpu") #Step 3: Run
#
# And you can print the results:
# ``print(results)``
#

################################################################################
# Step 1.5: Tune [Optional & Recommended]
# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#
# Run speed can further be improved by tuning. This optional step uses
# machine learning to look at each operation within a model (a function) and
# tries to find a faster way to run it. We do this through a cost model, and
# benchmarking possible schedules.
#
# The target is the same as compile.
#
# .. code-block:: python
#
# tvmc.tune(model, target="llvm") #Step 1.5: Optional Tune
#
# The terminal output should look like:
#
# .. code-block:: python
#
# [Task 1/13] Current/Best: 82.00/ 106.29 GFLOPS | Progress: (48/769) | 18.56 s
# [Task 1/13] Current/Best: 54.47/ 113.50 GFLOPS | Progress: (240/769) | 85.36 s
# .....
#
# There may be UserWarnings that can be ignored.
# This should make the end result faster, but it can take hours to tune.
#
# See the section 'Saving the Tuning Results' below. Be sure to pass the tuning
# results into compile if you want the results to apply.
#
# .. code-block:: python
#
# #tvmc.compile(model, target="llvm", tuning_records = "records.log") #Step 2: Compile

################################################################################
# Save and then start the process in the terminal:
# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#
# .. code-block:: python
#
# python my_tvmc_script.py
#
# Note: Your fans may become very active
#

################################################################################
# Example results:
# ~~~~~~~~~~~~~~~~
#
# .. code-block:: python
#
# Time elapsed for training: 18.99 s
# Execution time summary:
# mean (ms) max (ms) min (ms) std (ms)
# 25.24 26.12 24.89 0.38
#
#
# Output Names:
# ['output_0']
#


################################################################################
# Additional TVMC Functionalities
# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#

################################################################################
# Saving the model
# ~~~~~~~~~~~~~~~~
#
# To make things faster for later, after loading the model (Step 1) save the Relay version.
# The model will then appear where you saved it for later in the coverted syntax.
#
# .. code-block:: python
#
# model = tvmc.load('my_model.onnx') #Step 1: Load
# model.save(desired_model_path)
#
#

################################################################################
# Saving the package
# ~~~~~~~~~~~~~~~~~~
#
# After the model has been compiled (Step 2) the package also is also saveable.
#
# .. code-block:: python
#
# tvmc.compile(model, target="llvm", package_path="whatever")
#
# new_package = tvmc.TVMCPackage(package_path="whatever")
# result = tvmc.run(new_package) #Step 3: Run
#
#

################################################################################
# Using Autoscheduler
# ~~~~~~~~~~~~~~~~~~~
#
# Use the next generation of tvm to enable potentially faster run speed results.
# The search space of the schedules is automatically generated unlike
# previously where they needed to be hand written. (Learn more:
# `1 <https://tvm.apache.org/2021/03/03/intro-auto-scheduler>`_,
# `2 <https://arxiv.org/abs/2006.06762>`_)
#
# .. code-block:: python
#
# tvmc.tune(model, target="llvm", enable_autoscheduler = True)
#
#

################################################################################
# Saving the tuning results
# ~~~~~~~~~~~~~~~~~~~~~~~~~
#
# The tuning results can be saved in a file for later reuse.
#
# Method 1:
# .. code-block:: python
#
# log_file = "hello.json"
#
# # Run tuning
# tvmc.tune(model, target="llvm",tuning_records=log_file)
#
# ...
#
# # Later run tuning and reuse tuning results
# tvmc.tune(model, target="llvm",tuning_records=log_file)
#
# Method 2:
# .. code-block:: python
#
# # Run tuning
# tuning_records = tvmc.tune(model, target="llvm")
#
# ...
#
# # Later run tuning and reuse tuning results
# tvmc.tune(model, target="llvm",tuning_records=tuning_records)
#

################################################################################
# Tuning a more complex model:
# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#
# If you notice T's printing that look like ``.........T.T..T..T..T.T.T.T.T.T.``
# increase the searching time frame:
#
# .. code-block:: python
#
# tvmc.tune(model,trials=10000,timeout=10,)
#

################################################################################
# Compiling a model for a remote device:
# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#
# A remote procedural call (RPC) is useful when you would like to compile for hardware
# that is not on your local machine. The tvmc methods support this.
# To set up the RPC server take a look at the 'Set up RPC Server on Device'
# section in this `document <https://tvm.apache.org/docs/tutorials/get_started/cross_compilation_and_rpc.html>`_.
#
# Within the TVMC Script include the following and adjust accordingly:
#
# .. code-block:: python
#
# tvmc.tune(
# model,
# target=target, # Compilation target as string // Device to compile for
# target_host=target_host, # Host processor
# hostname=host_ip_address, #The IP address of an RPC tracker, used when benchmarking remotely.
# port=port_number, # The port of the RPC tracker to connect to. Defaults to 9090.
# rpc_key=your_key, # The RPC tracker key of the target device. Required when rpc_tracker is provided
# )
#