You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Here, we purpose several steps in developing Paddle API, make developing progress smooth and acceptable.
Use SWIG-API to expose Paddle and decide Python API.
from 2016-12-20 to 2016-12-27
give an example how to train a model in Python currently. @reyoung [2016-12-20]
MNIST demo and Quick Start Demo.
local mode / cluster mode.
make trainer_config & data_provider in a same file. [2016-12-21, 22, 23]
make trainer_config in the same python file. @jacquesqiao
make data can easily fit to paddle gradient machine
extract user level API from demo [2016-12-24,25]
Add some syntactic sugar in python, make users not to write some standard code.
change other demos by using API trainer. [2016-12-26,27]
Make Python-API stable, change it to C-API.
We hope the Python-API is in alpha stage in 2016-12-27. So we can discuss API design base on this version. We will decide what Python API should be in early January.
When the python API about Paddle is stable, we will remove SWIG and change it to C-API, let multiple language bindings could be possible.
1. Clean Global variables/Remove log fatal.
1. Expose C-API and make Python-API stable
Releasing C-API and SDK for C/Cpp model inference.
The text was updated successfully, but these errors were encountered:
* test=develop, update install doc
* add setuptools dependency
* Update compile_Ubuntu.md
update cudnn version
* Update install_Ubuntu.md
update cudnn version
* Update install_CentOS.md
update cudnn version
* test=develop, refine ubuntu cuda support
* test=develop, refine ubuntu cuda support
* test=develop, refine python support on python3.5
* fix python2 in ubuntu16.04 need gcc 5.4
* add nccl related doc
* refine doc
* add gpu fast install doc
* refine command of pip
* add avx off for mac complie
* refine code
* refine doc
* fix python only support 64bit
Here, we purpose several steps in developing Paddle API, make developing progress smooth and acceptable.
Use SWIG-API to expose Paddle and decide Python API.
from 2016-12-20 to 2016-12-27
Make Python-API stable, change it to C-API.
1. Clean Global variables/Remove log fatal.
1. Expose C-API and make Python-API stable
Releasing C-API and SDK for C/Cpp model inference.
The text was updated successfully, but these errors were encountered: