Skip to content

P
pytensor
提交 ad23a259 authored 作者: notoraptor's avatar notoraptor
浏览文件
操作

Prepare release candidate 1.0.0rc1

上级 184216ae
隐藏空白字符变更
内嵌 并排
正在显示 包含 139 行增加 和 33 行删除
HISTORY.txt
浏览文件 @ ad23a259
...@@ -5,6 +5,119 @@ ...@@ -5,6 +5,119 @@
Old Release Notes Old Release Notes
================= =================
=============
Release Notes
=============
Theano 0.10.0beta4 (16th of October, 2017)
==========================================
This release contains new features, improvements and bug fixes to prepare the upcoming release candidate.
We recommend that every developer updates to this version.
Highlights:
- Announcing that `MILA will stop developing Theano <https://groups.google.com/d/msg/theano-users/7Poq8BZutbY/rNCIfvAEAwAJ>`_
- Bug fixes, crash fixes, warning improvements and documentation updates
A total of 70 people contributed to this release since 0.9.0, see list below.
Interface changes:
- Generalized ``AllocDiag`` for any non-scalar input
Convolution updates:
- Implemented fractional bilinear upsampling
cuDNN (GPU):
- Disallowed ``float16`` precision for convolution gradients
- Fixed memory alignment detection
- Added profiling in C debug mode (with theano flag ``cmodule.debug=True``)
New features:
- Implemented truncated normal distribution with box-muller transform
- Added ``L_op()`` overriding option for ``OpFromGraph``
- Added NumPy C-API based fallback implementation for ``[sd]gemv_`` and ``[sd]dot_``
Other more detailed changes:
- Improved stack trace follow-up for GPU optimizations
- Fixed gradient error for elemwise ``minimum`` and ``maximum`` when compared values are the same
- Fixed gradient for ``ARange``
- Removed ``ViewOp`` subclass during optimization
Commiters since 0.9.0:
- Frederic Bastien
- João Victor Tozatti Risso
- Arnaud Bergeron
- Steven Bocco
- Mohammed Affan
- amrithasuresh
- Pascal Lamblin
- Reyhane Askari
- Alexander Matyasko
- Shawn Tan
- Simon Lefrancois
- Adam Becker
- Vikram
- Gijs van Tulder
- Faruk Ahmed
- Thomas George
- erakra
- Andrei Costinescu
- Boris Fomitchev
- Zhouhan LIN
- Aleksandar Botev
- jhelie
- xiaoqie
- Tegan Maharaj
- Matt Graham
- Cesar Laurent
- Gabe Schwartz
- Juan Camilo Gamboa Higuera
- Tim Cooijmans
- Anirudh Goyal
- Saizheng Zhang
- Yikang Shen
- vipulraheja
- Florian Bordes
- Sina Honari
- Chiheb Trabelsi
- Shubh Vachher
- Daren Eiri
- Joseph Paul Cohen
- Laurent Dinh
- Mohamed Ishmael Diwan Belghazi
- Jeff Donahue
- Ramana Subramanyam
- Bogdan Budescu
- Dzmitry Bahdanau
- Ghislain Antony Vaillant
- Jan Schlüter
- Nan Jiang
- Xavier Bouthillier
- fo40225
- mrTsjolder
- wyjw
- Aarni Koskela
- Adam Geitgey
- Adrian Keet
- Adrian Seyboldt
- Anmol Sahoo
- Chong Wu
- Holger Kohr
- Jayanth Koushik
- Lilian Besson
- Lv Tao
- Michael Manukyan
- Murugesh Marvel
- NALEPA
- Zotov Yuriy
- dareneiri
- lrast
- morrme
- naitonium
Theano 0.10.0beta3 (20th of September, 2017) Theano 0.10.0beta3 (20th of September, 2017)
============================================ ============================================
......
NEWS.txt
浏览文件 @ ad23a259
...@@ -3,46 +3,26 @@ Release Notes ...@@ -3,46 +3,26 @@ Release Notes
============= =============
Theano 0.10.0beta4 (16th of October, 2017) Theano 1.0.0rc1 (30th of October, 2017)
========================================== =======================================
This release contains new features, improvements and bug fixes to prepare the upcoming release candidate. This release contains new features, improvements and bug fixes to prepare the upcoming release.
We recommend that every developer updates to this version. We recommend that every developer updates to this version.
Highlights: Highlights:
- Announcing that `MILA will stop developing Theano <https://groups.google.com/d/msg/theano-users/7Poq8BZutbY/rNCIfvAEAwAJ>`_ - Make sure MKL uses GNU OpenMP
- Bug fixes, crash fixes, warning improvements and documentation updates - Optimized ``SUM(x^2)``, ``SUM(ABS(X))`` and ``MAX(ABS(X))`` operations with cuDNN reductions
- Added Python scripts to help test cuDNN convolutions
- Fixed invalid casts and index overflows in ``theano.tensor.signal.pool``
A total of 70 people contributed to this release since 0.9.0, see list below. A total of 71 people contributed to this release since 0.9.0, see list below.
Interface changes:
- Generalized ``AllocDiag`` for any non-scalar input
Convolution updates:
- Implemented fractional bilinear upsampling
cuDNN (GPU):
- Disallowed ``float16`` precision for convolution gradients
- Fixed memory alignment detection
- Added profiling in C debug mode (with theano flag ``cmodule.debug=True``)
New features:
- Implemented truncated normal distribution with box-muller transform
- Added ``L_op()`` overriding option for ``OpFromGraph``
- Added NumPy C-API based fallback implementation for ``[sd]gemv_`` and ``[sd]dot_``
Other more detailed changes:
- Improved stack trace follow-up for GPU optimizations
- Fixed gradient error for elemwise ``minimum`` and ``maximum`` when compared values are the same
- Fixed gradient for ``ARange``
- Removed ``ViewOp`` subclass during optimization
Commiters since 0.9.0: Commiters since 0.9.0:
- Frederic Bastien - Frederic Bastien
- Steven Bocco
- João Victor Tozatti Risso - João Victor Tozatti Risso
- Arnaud Bergeron - Arnaud Bergeron
- Steven Bocco
- Mohammed Affan - Mohammed Affan
- amrithasuresh - amrithasuresh
- Pascal Lamblin - Pascal Lamblin
...@@ -104,6 +84,7 @@ Commiters since 0.9.0: ...@@ -104,6 +84,7 @@ Commiters since 0.9.0:
- Michael Manukyan - Michael Manukyan
- Murugesh Marvel - Murugesh Marvel
- NALEPA - NALEPA
- Rebecca N. Palmer
- Zotov Yuriy - Zotov Yuriy
- dareneiri - dareneiri
- lrast - lrast
......
NEWS_DEV.txt
浏览文件 @ ad23a259
...@@ -5,7 +5,7 @@ DRAFT Release Notes ...@@ -5,7 +5,7 @@ DRAFT Release Notes
=================== ===================
git log -p rel-0.9.0... |grep Merge|grep '#[0123456789]' |cut -f 8 -d ' ' | sed 's\#\* https://github.com/Theano/Theano/pull/\' git log -p rel-0.9.0... |grep Merge|grep '#[0123456789]' |cut -f 8 -d ' ' | sed 's\#\* https://github.com/Theano/Theano/pull/\'
git log -p rel-0.10.0beta3... |grep Merge|grep '#[0123456789]' |cut -f 8 -d ' ' | sed 's\#\* https://github.com/Theano/Theano/pull/\' git log -p rel-0.10.0beta4... |grep Merge|grep '#[0123456789]' |cut -f 8 -d ' ' | sed 's\#\* https://github.com/Theano/Theano/pull/\'
# Commit count per user # Commit count per user
git shortlog -sn rel-0.9.0.. git shortlog -sn rel-0.9.0..
...@@ -30,6 +30,7 @@ Highlights: ...@@ -30,6 +30,7 @@ Highlights:
- Theano now internally uses ``sha256`` instead of ``md5`` to work on systems that forbide ``md5`` for security reason - Theano now internally uses ``sha256`` instead of ``md5`` to work on systems that forbide ``md5`` for security reason
- Removed old GPU backend ``theano.sandbox.cuda``. New backend ``theano.gpuarray`` is now the official GPU backend - Removed old GPU backend ``theano.sandbox.cuda``. New backend ``theano.gpuarray`` is now the official GPU backend
- Support more debuggers for ``PdbBreakpoint`` - Support more debuggers for ``PdbBreakpoint``
- Make sure MKL uses GNU OpenMP
- Improved elemwise operations - Improved elemwise operations
...@@ -107,12 +108,13 @@ GPU: ...@@ -107,12 +108,13 @@ GPU:
- Better support and loading on Windows and Mac - Better support and loading on Windows and Mac
- Support cuDNN v6 dilated convolutions - Support cuDNN v6 dilated convolutions
- Support cuDNN v6 reductions for contiguous inputs - Support cuDNN v6 reductions for contiguous inputs
- Optimized ``SUM(x^2)``, ``SUM(ABS(X))`` and ``MAX(ABS(X))`` operations with cuDNN reductions
- Added new Theano flags ``cuda.include_path``, ``dnn.base_path`` and ``dnn.bin_path`` - Added new Theano flags ``cuda.include_path``, ``dnn.base_path`` and ``dnn.bin_path``
to help configure Theano when CUDA and cuDNN can not be found automatically. to help configure Theano when CUDA and cuDNN can not be found automatically.
- Disallowed ``float16`` precision for convolution gradients - Disallowed ``float16`` precision for convolution gradients
- Fixed memory alignment detection - Fixed memory alignment detection
- Added profiling in C debug mode (with theano flag ``cmodule.debug=True``) - Added profiling in C debug mode (with theano flag ``cmodule.debug=True``)
- Added Python scripts to help test cuDNN convolutions
- Updated ``float16`` support - Updated ``float16`` support
...@@ -155,6 +157,7 @@ Others: ...@@ -155,6 +157,7 @@ Others:
- Added decsription for rnnblock - Added decsription for rnnblock
Other more detailed changes: Other more detailed changes:
- Fixed invalid casts and index overflows in ``theano.tensor.signal.pool``
- Fixed gradient error for elemwise ``minimum`` and ``maximum`` when compared values are the same - Fixed gradient error for elemwise ``minimum`` and ``maximum`` when compared values are the same
- Fixed gradient for ``ARange`` - Fixed gradient for ``ARange``
- Removed ``ViewOp`` subclass during optimization - Removed ``ViewOp`` subclass during optimization
...@@ -170,6 +173,13 @@ Other more detailed changes: ...@@ -170,6 +173,13 @@ Other more detailed changes:
- Many improvements for Jenkins CI tests: daily testings on Mac and Windows in addition to Linux - Many improvements for Jenkins CI tests: daily testings on Mac and Windows in addition to Linux
ALL THE PR BELLOW HAVE BEEN CHECKED ALL THE PR BELLOW HAVE BEEN CHECKED
* https://github.com/Theano/Theano/pull/6496
* https://github.com/Theano/Theano/pull/6495
* https://github.com/Theano/Theano/pull/6492
* https://github.com/Theano/Theano/pull/6489
* https://github.com/Theano/Theano/pull/6488
* https://github.com/Theano/Theano/pull/6490
* https://github.com/Theano/Theano/pull/5932
* https://github.com/Theano/Theano/pull/6479 * https://github.com/Theano/Theano/pull/6479
* https://github.com/Theano/Theano/pull/6401 * https://github.com/Theano/Theano/pull/6401
* https://github.com/Theano/Theano/pull/6472 * https://github.com/Theano/Theano/pull/6472
......
doc/index.txt
浏览文件 @ ad23a259
...@@ -21,6 +21,8 @@ learning/machine learning <https://mila.umontreal.ca/en/cours/>`_ classes). ...@@ -21,6 +21,8 @@ learning/machine learning <https://mila.umontreal.ca/en/cours/>`_ classes).
News News
==== ====
* 2017/10/30: Release of Theano 1.0.0rc1, new features and many bugfixes, final release to coming.
* 2017/10/16: Release of Theano 0.10.0beta4, new features and many bugfixes, release candidate to coming. * 2017/10/16: Release of Theano 0.10.0beta4, new features and many bugfixes, release candidate to coming.
* 2017/09/28: IMPORTANT: `MILA will stop developing Theano <https://groups.google.com/d/msg/theano-users/7Poq8BZutbY/rNCIfvAEAwAJ>`_ and the next release (renamed to 1.0) will be the last main release. * 2017/09/28: IMPORTANT: `MILA will stop developing Theano <https://groups.google.com/d/msg/theano-users/7Poq8BZutbY/rNCIfvAEAwAJ>`_ and the next release (renamed to 1.0) will be the last main release.
......
doc/introduction.txt
浏览文件 @ ad23a259
...@@ -165,7 +165,7 @@ Note: There is no short term plan to support multi-node computation. ...@@ -165,7 +165,7 @@ Note: There is no short term plan to support multi-node computation.
Theano Vision State Theano Vision State
=================== ===================
Here is the state of that vision as of October 16th, 2017 (after Theano 0.10.0beta4): Here is the state of that vision as of October 30th, 2017 (after Theano 1.0.0rc1):
* `MILA will stop developing Theano. <https://groups.google.com/d/msg/theano-users/7Poq8BZutbY/rNCIfvAEAwAJ>`_ * `MILA will stop developing Theano. <https://groups.google.com/d/msg/theano-users/7Poq8BZutbY/rNCIfvAEAwAJ>`_
* We support tensors using the `numpy.ndarray` object and we support many operations on them. * We support tensors using the `numpy.ndarray` object and we support many operations on them.
......
theano/version.py
浏览文件 @ ad23a259
...@@ -2,7 +2,7 @@ from __future__ import absolute_import, print_function, division ...@@ -2,7 +2,7 @@ from __future__ import absolute_import, print_function, division
from theano._version import get_versions from theano._version import get_versions
FALLBACK_VERSION = "1.0.0dev+unknown" FALLBACK_VERSION = "1.0.0rc1+unknown"
info = get_versions() info = get_versions()
if info['error'] is not None: if info['error'] is not None:
......
Markdown 格式
0% 或
您添加了 0 人 到此讨论。请谨慎行事。
请先完成此评论的编辑！
请 注册 或者 后发表评论