Skip to content
Projects
Groups
Snippets
Help
This project
Loading...
Sign in / Register
Toggle navigation
T
tic
Overview
Overview
Details
Activity
Cycle Analytics
Repository
Repository
Files
Commits
Branches
Tags
Contributors
Graph
Compare
Charts
Issues
0
Issues
0
List
Board
Labels
Milestones
Merge Requests
0
Merge Requests
0
CI / CD
CI / CD
Pipelines
Jobs
Schedules
Charts
Wiki
Wiki
Snippets
Snippets
Members
Members
Collapse sidebar
Close sidebar
Activity
Graph
Charts
Create a new issue
Jobs
Commits
Issue Boards
Open sidebar
wenyuanbo
tic
Commits
f4a15334
Commit
f4a15334
authored
Apr 16, 2017
by
Tianqi Chen
Committed by
GitHub
Apr 16, 2017
Browse files
Options
Browse Files
Download
Email Patches
Plain Diff
[DOC] move comments to file header (#91)
parent
b8234498
Hide whitespace changes
Inline
Side-by-side
Showing
20 changed files
with
43 additions
and
37 deletions
+43
-37
README.md
+6
-2
docs/api/python/collection.rst
+0
-2
docs/api/python/dev.rst
+2
-2
docs/api/python/index.rst
+0
-1
docs/api/python/module.rst
+1
-6
docs/api/python/ndarray.rst
+1
-2
docs/api/python/schedule.rst
+1
-2
docs/api/python/tensor.rst
+1
-2
docs/api/python/tvm.rst
+1
-2
docs/conf.py
+5
-4
docs/genindex.rst
+2
-0
docs/index.rst
+2
-0
python/tvm/__init__.py
+1
-1
python/tvm/build.py
+4
-4
python/tvm/collections.py
+1
-1
python/tvm/module.py
+6
-1
python/tvm/ndarray.py
+3
-2
python/tvm/node.py
+4
-1
python/tvm/schedule.py
+1
-1
python/tvm/tensor.py
+1
-1
No files found.
README.md
View file @
f4a15334
...
...
@@ -8,5 +8,9 @@
[
Release Notes
](
NEWS.md
)
# TVM
TVM is a domain specific language(DSL) for tensor computations.
The goal of the project is to generate efficient kernels for deep learning workloads.
TVM is a low level domain specific language(DSL) for compiling tensor computation pipelines.
It is designed to compile multi-dimensional tensor algebra pipelines which
are crucial to deep learning frameworks.
## Documentation
The current documentation can be build locally via sphinx. See
[
docs
](
docs
)
folder for details.
docs/api/python/collection.rst
View file @
f4a15334
tvm.collections
---------------
Collections contains data structures used in TVM DSL.
.. automodule:: tvm.collections
:members:
docs/api/python/dev.rst
View file @
f4a15334
Developer API
-------------
This page contains modules that are used by developers of TVM.
Many of these APIs are PackedFunc registered in C++ backend.
tvm.node
~~~~~~~~
Node is the base class of all TVM AST. Normally user do not
need to touch this api.
.. automodule:: tvm.node
.. autoclass:: tvm.node.NodeBase
:members:
...
...
docs/api/python/index.rst
View file @
f4a15334
...
...
@@ -11,6 +11,5 @@ Python API
module
ndarray
collection
node
function
dev
docs/api/python/module.rst
View file @
f4a15334
tvm.module
----------
.. auto
class:: tvm.module.M
odule
.. auto
module:: tvm.m
odule
:members:
:inherited-members:
.. autofunction:: tvm.module.load
.. autofunction:: tvm.module.enabled
docs/api/python/ndarray.rst
View file @
f4a15334
tvm.ndarray
-----------
tvm.ndarray provides a minimum runtime array API to testing out
the correctness of the program.
.. automodule:: tvm.ndarray
.. autoclass:: tvm.ndarray.TVMContext
:members:
...
...
docs/api/python/schedule.rst
View file @
f4a15334
tvm.schedule
------------
The `tvm.schedule` module contains classes of scheduling
structure of tvm.
.. automodule:: tvm.schedule
.. autoclass:: tvm.schedule.IterVar
:members:
...
...
docs/api/python/tensor.rst
View file @
f4a15334
tvm.tensor
----------
The `tvm.tensor` module contains declaration of Tensor
and Operation class for computation declaration.
.. automodule:: tvm.tensor
.. autoclass:: tvm.tensor.Tensor
:members:
...
...
docs/api/python/tvm.rst
View file @
f4a15334
tvm
---
tvm is a library root namespace contains functions for
declaring computation.
The user facing API for computation declaration.
.. autosummary::
...
...
docs/conf.py
View file @
f4a15334
...
...
@@ -156,7 +156,10 @@ latex_documents = [
def
run_doxygen
(
folder
):
"""Run the doxygen make command in the designated folder."""
try
:
retcode
=
subprocess
.
call
(
"cd
%
s; make doc"
%
folder
,
shell
=
True
)
#retcode = subprocess.call("cd %s; make doc" % folder, shell=True)
retcode
=
subprocess
.
call
(
"rm -rf _build/html/doxygen"
,
shell
=
True
)
retcode
=
subprocess
.
call
(
"mkdir -p _build/html"
,
shell
=
True
)
retcode
=
subprocess
.
call
(
"cp -rf doxygen/html _build/html/doxygen"
,
shell
=
True
)
if
retcode
<
0
:
sys
.
stderr
.
write
(
"doxygen terminated by signal
%
s"
%
(
-
retcode
))
except
OSError
as
e
:
...
...
@@ -174,9 +177,7 @@ gallery_dirs = ['tutorials']
def
generate_doxygen_xml
(
app
):
"""Run the doxygen make commands if we're on the ReadTheDocs server"""
read_the_docs_build
=
os
.
environ
.
get
(
'READTHEDOCS'
,
None
)
==
'True'
if
read_the_docs_build
:
run_doxygen
(
'..'
)
run_doxygen
(
'..'
)
def
setup
(
app
):
# Add hook for building doxygen xml when needed
...
...
docs/genindex.rst
0 → 100644
View file @
f4a15334
Index
=====
docs/index.rst
View file @
f4a15334
...
...
@@ -9,7 +9,9 @@ Contents
.. toctree::
:maxdepth: 1
self
how_to/install
tutorials/index
api/python/index
how_to/contribute
genindex
python/tvm/__init__.py
View file @
f4a15334
# pylint: disable=redefined-builtin, wildcard-import
"""
C++ backend related python scripts
"""
"""
TVM: a DSL for tensor kernel compilation
"""
from
__future__
import
absolute_import
as
_abs
from
.
import
tensor
...
...
python/tvm/build.py
View file @
f4a15334
"""The build
pipeline
in python.
"""The build
utils
in python.
Eventually some of these pipelines will be moved to C++.
But the first pipeline will be kept in python for ease of change and evolving
.
This module provides the functions to transform schedule to
LoweredFunc and compiled Module
.
"""
from
__future__
import
absolute_import
as
_abs
from
.
import
api
from
.
import
tensor
from
.
import
schedule
...
...
python/tvm/collections.py
View file @
f4a15334
"""Collection
structure in the high level
DSL."""
"""Collection
s contains data structures used in TVM
DSL."""
from
__future__
import
absolute_import
as
_abs
from
._ctypes._node
import
NodeBase
,
register_node
from
.
import
_api_internal
...
...
python/tvm/module.py
View file @
f4a15334
"""
Runtime module related stuffs
"""
"""
Container of compiled functions of TVM.
"""
from
__future__
import
absolute_import
as
_abs
from
._ctypes._function
import
ModuleBase
,
_init_module_module
from
._ctypes._function
import
_init_api
...
...
@@ -21,6 +21,11 @@ class Module(ModuleBase):
----------
fmt : str, optional
The specified format.
Returns
-------
source : str
The result source code.
"""
return
_GetSource
(
self
,
fmt
)
...
...
python/tvm/ndarray.py
View file @
f4a15334
"""TVM Runtime API.
"""TVM Runtime
NDArray
API.
This is a simplified runtime API for quick testing and proptyping.
tvm.ndarray provides a minimum runtime array API to test
the correctness of the program.
"""
# pylint: disable=invalid-name,unused-import
from
__future__
import
absolute_import
as
_abs
...
...
python/tvm/node.py
View file @
f4a15334
"""Namespace for base node class"""
"""Node is the base class of all TVM AST.
Normally user do not need to touch this api.
"""
# pylint: disable=unused-import
from
__future__
import
absolute_import
as
_abs
from
._ctypes._node
import
NodeBase
,
register_node
...
...
python/tvm/schedule.py
View file @
f4a15334
"""
Collection structure in the high level DSL
."""
"""
The computation schedule api of TVM
."""
from
__future__
import
absolute_import
as
_abs
from
._ctypes._node
import
NodeBase
,
register_node
from
.
import
_api_internal
...
...
python/tvm/tensor.py
View file @
f4a15334
"""Tensor and
Computation abstraction objects
"""
"""Tensor and
Operation class for computation declaration.
"""
# pylint: disable=invalid-name
from
__future__
import
absolute_import
as
_abs
from
._ctypes._node
import
NodeBase
,
SliceBase
,
register_node
,
convert_to_node
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment