* Provide English tutorial on how to use VisualDL in Keras

* Add the English Tutorial of PyTorch.

* Add the English MXNet tutorial.

* Remove links in the MXNet tutorial.

* Add the support to write documentation in the pybind directly.

* Provide more detail of the API documentations.

* Add Python 3 support

* Update protobuf to 3.5 and pybind to 2.2.2

* Update test scripts to use python -m

* Change zip/xrange/import usages

* Remove file generated by

* Graph viz set up (#282)

* Include the Graph requirement, GraphViz` in docs.

* Provide information to the ONNX format model export.

* Ask the user to download Graphviz via the website instead of brew.

* Make test and build scripts compatible with python 2 & 3

* A WITH_PYTHON3 option was added to cmake.

* setup.py will detect runing version python, and setup up WITH_PYTHON3 environemnt variable before calling build.sh

* build.sh and test.sh will handle WITH_PYTHON3

* Fix pybind warning

* Pass WITH_PYTHON3 correctly

* Fix coding style

* Add six for Python 2 and 3 Compatibility

* Fix coding style

Update the link in how_to_dev_backend_en.md to the full path. This should resolve the issue where the API doc not building. (#288)

* Include the Graph requirement, GraphViz` in docs.

* Provide information to the ONNX format model export.

* Ask the user to download Graphviz via the website instead of brew.

* Adds backend / SDK dev doc

* Update per Varun and chanwei's comment

* rename vdl_scratch.py to vdl_create_scratch_log

* Simply replace the vdl_scratch.py

* Binary Record Fixes

* Fixed issues with binary_record.h filename generator method.  std::hash is not consistent across OS.  It will generate different hash's when ran on different OS on the same string.  Change the filename generator to use a static counter.  It will be more ideal to use UUID generator available in boost 1.4.2, however, boost 1.4.2 has problems compiling on c++11 \n* Fix issue with pytorch demo where it logs to the absolute path /workspace

* Fix performance issue on server when server fails to load image file.  Currently server retries an API call 3 times before it fails, at which point it'll sleep the thread for 2 secs.  This becomes problematic when there are alot of SDK errors.  I'll create seperate issue to refine the retry logic to not be so expensive on the server.

* Provide instructions on virtualenv.

* Clean up the detail a little bit.

* Fix python path typo in the start dev server script
Kill webpack in case of python error.

* Switch to use trap

* capture more signal.

VDL is set to retry an pybinded API call multiple times in case the call fails the first time.  The issue is the method lib.retry will sleep the thread for 2 seconds after an exception.  Since VDL was not running in threaded mode, this causes all request to the server to stop immediately, therefore even requests to get static resource are frozen.  This change enables threaded mode in flask. (#249)

* Add how to develop frontend documentation.

* Provide a bit of information on san file.

* fix some information error

* Update the documention.

* Simplify the documentation.

* Remove unnecessary sentences

Add script to call webpack to deploy and watch frontend code from within flask.  Optimize script location (#236)

* Add script to call webpack to deploy and watch frontend code from within flask.  Optimize script location

* remove npm install

* Create "setup_dev_env.sh" script that will set up all the node and SDK depenencies.  Create "start_dev_server.sh" to start up dev server, which will start backend server and start webpack --watch on frontend.

* Add travis-ci job to deploy docs to visualdl.paddlepaddle.org.  Update readme's to correctly show all images.

* Add SSH known hosts

* Add in additional restrictions for documentation deploy:
* Only deploy on merge to 'develop' branch

Update the requirements.txt to include pre-commit

* Add the Chinese and English "Contribute Code" documentations.

* Provide some more detail messages in the documentations

* Add C++ and python style check to travis.  
* Update all C++/python code that violate coding standards.

Fix some typos

* Update README.md

Fix typo in the README python SDK example.

* Fix the typo in readme.cn

docker build -f Dockerfile.demo -t visualdl:demo .
docker run -p 8040:8040 visualdl:demo
docker run -p 8040:8040 -v <YOUR_LOG_DIR>:/scratch_log visualdl:demo

* fix build problem that the first build has no js.
* use for loop to check if frontend file is duplicated
* package_data use *.js *.html to match file

* fix #176