* test=develop, fix docker with paddle nccl problem

* test=develop, Add Variable api and refine dygraph related API

* test=develop, Add Variable api and refine dygraph related API

* test=develop, refine test for new api and error info

* test=develop, refine error info and test_layers

* test=develop, add API.spec

* test=devleop, fix to_string python2 and python3 compat error and refien doc

* test=devleop, add API spec

* test=devleop, update API spec

* test=devleop, update API spec

* test=develop, invoke ci

* test=develop, fix example code

* test=develop, update API spec

* test=develop, fix auto_prune_error_on_leaf

* test=develop, fix auto prune error on loss stop_gradient

* test=develop, remove useless error check

* test=develop, add more ut for sorted gradient

* Update crf_decoding api & example

* Update api spec

* Fix linear chain crf api

* Avoid sharing data pointer with input

* Simplify the logic in linear_chain_crf_decoding

* Add unittest for crf_decoding when label & path both are set

* Update API spec

* Add unittest for layers && correct infer_shape in chunk_eval

test=release/1.6

test=release/1.6

* Add multihead op for ernie opt

* Refine softmax

* Refine kernel.

* Refine cuda kernel

* Refine cuda version

* Refine cmake

1.support asymmetric padding;
2.support padding algorithm:"SAME" and "VALID";
3.support channel_last: data_format NHWC and NDHWC;
4.change doc of python API and c++;

test=release/1.6

* refector geo sgd & communicator

* add a default value for _prune interface

* modify document

* How to write custom op needs to follow framework OP spec.
* Package fluid_framework.so and headers into whl.
* Add paddle.sysconfig.get_include() and paddle.sysconfig.get_lib() to get include dir and lib dir.
* Export some C-APIs to merge OpInfo between core.so and custom_op.so.
* Add unit testing.
* Update API.spec.

* test=develop, argument shape support tensor and tensor in list

* test=develop,Increasing the coverage of CI tests

* test=develop, modify the document and update API.spec

* test=develop, modify the doc and update API.spec

* test=develop, modify the doc and update API.spec

* test=develop, modify the interface of UniformInitializer

* test=develop, modify the interface of XavierInitializer and MSRAInitializer

* test=develop, modify based on review's comments

* test=develop, modify based on review's comments

*  test=develop, modify based on review's comments

* fix pool2d pool3d:
1. support asymmetric padding;
2. support padding algorithm:"SAME" and "VALID";
3. support channel_last: data_format NHWC and NDHWC;
4. support inferring shape when input with negative dims in compile time;
5. change doc of python API and c++;
6. fix bug in cuda kernel when Attr(adaptive) is true.

test=develop,test=document_preview

* fix 'tensors' to 'Tensors'. test=develop,test=document_preview

* add test for converage ValueError.test=develop,test=document_preview

* resolve conflict in test_pool2d. test=develop

* opt FC

* opt rest of dygraph.nn

* new param shape check and unittest

* add kwargs for Layer

* add new set_value api

* use property decorator

* update API.spec, test=develop

* use UserList, separate gettersetters, test=develop

* update test_custom_layer_with_kwargs, test=develop

* fix UserList compatibility, test=develop

* fix UserList compatibility, test=develop

* keep FC._w, test=develop

* add unittests, Conv3D bug fix, test=develop

* clean code, test=develop

* fix dygraph guard in unittest, test=develop

* add property setters, remove unused param in tracer, test=develop

* tracer none check, test=develop

* merge, test=develop

* refine, test=develop

* bug fix in  prelu and conv3d_transpose, test=develop

* rm __set__, test=develop

* set tensor value instead of assign op

* fix property setter call, test=develop

* fix api.spec, test=develop

* fix doc sample, test=develop

* add parameter shape check when load parameter from file; test=develop

* fix test error; test=develop

* add wrong shape check; test=develop

* remove useless code; test=develop

* add testcase setup

* add teardown, remove temp model_path; test=develop

* add clean process; test=develop

* test=develop, fix docker with paddle nccl problem

* test=develop, Add Variable api and refine dygraph related API

* test=develop, Add Variable api and refine dygraph related API

* test=develop, refine test for new api and error info

* test=develop, refine error info and test_layers

* test=develop, add API.spec

* test=devleop, fix to_string python2 and python3 compat error and refien doc

* test=devleop, add API spec

* test=devleop, update API spec

* test=devleop, update API spec

* test=develop, invoke ci

* test=develop, fix example code

* test=develop, update API spec

* test=develop, add compat test and fix inplace campat dict error

* fix pool infershape. test=develop

* fix unittest converage. test=develop

* fix format. test=develop

* make pad and split support fp16
test=develop

* add a base class for the Communicator
* add AsyncCommunicator Impl for async distributed training

test=develop
Signed-off-by: Nzhaoyuchen <zhaoyuchen01@baidu.com>

* impove error message when passing ndarray with object dtype

* imporve message format

* change assert to raise TypeError

* remind user how to locate the irregular data instead of printing

* add unittest for input array type check

* Fix conv2d+dequantize squash for residual fusion

test=develop

* Correct int8 input

test=develop

* Add if exclude or include padding in pool2d mkldnn

test=develop

* x.dims == y.dims test=develop

* refine comment

* Expose `mutable_data` as python binding

test=develop

* Add test for device pointer binding

test=develop

* Make test compatible with python 2

* fix input shape check test=develop

* move PADDLE_ENFORCE test=develop

Add dtype for coalesce_tensor_op

The new "fluid.data" changes old "fluid.layers.data":

1. Add shape and dtype check.
2. Remove "append_batch_size" parameter. We won't offer this in the new data layer because other deep learning platforms don't have this kind of data layer pre-processing. It may confuse users.
3. Remove "stop gradient" parameter because the data layer doesn't do back-propagation

TODO：
Now data layer feeded by executor is checked, will we want to check the feed data of readers in the future?

* add kernel for fill_op, test=develop

* modify PADDLE_ENFORCE to PADDLE_ENFORCE_EQ, test=develop

* add op test for fill_op, test=develop

* REGISTER COP CUDA KERNEL, test=develop

* update test_fill_op.py, test=develop

* change FillConstantOpVarTypeInference to FillOpVarTypeInference, test=develop

* fix op test, test=develop

* add head file, test=develop

* add support tensor and tensorlist for strided_slice OP test=develop

* fix the commnet test=develop

* fix test=develop

* fix the bug test=develop

* delete log test=develop

* fix API.spec test=develop

* fix test=develop

* fix bn

* treat broadcast as non-initial, test=develop

* rename the class name

* rename the class name, test=develop

* add support of matmul with multiple head even different width and height

Original matmul with multiple head supports only the mat_a.width == mat_b.height,
in that case, mat_b will be horizontally split. In this patch, we extend the
support when mat_a.width != mat_b.height but mat_a.width/head_number == mat_b.height,
in this case, mab_b will be vertically split.

One example is A is [3, 8], B is [2, 16], head_number is 4. In this
case, A will be split as [3, 2], B will be (vertically) split as
[2, 4]. The final result will be 4 matrix of 4 matrix of [3,4], i.e. [3, 16]

test=develop

* add support of matmul with multiple head even different width and height

Original matmul with multiple head supports only the mat_a.width == mat_b.height,
in that case, mat_b will be horizontally split. In this patch, we extend the
support when mat_a.width != mat_b.height but mat_a.width/head_number == mat_b.height,
in this case, mab_b will be vertically split.

One example is A is [3, 8], B is [2, 16], head_number is 4. In this
case, A will be split as [3, 2], B will be (vertically) split as
[2, 4]. The final result will be 4 matrix of 4 matrix of [3,4], i.e. [3, 16]

test=develop

* refactor the code of matmul with multiple head even different width and height

test=develop

* refine ctc align op with padding 
* refine api sample code

* add input type and dtype check for softmax_op

test=develop

* refine error message

test=develop

* Removing last dims constraints of seq_pad and seq_unpad test=develop

* fix test_layer api code test=develop

* fix sequence_pad_op.cc conflict test=develop

* remove test_analyzer_mm_dnn test=develop

* fix vectorize bug test=develop

* fix vectorize<int> test=develop

* Add float16 support to `sync_batch_norm_op`

test=develop

* Add test for sync_bn with FP16 input

test=develop

* Remove constraint that last dimension is forced to be 1 by add
lookup_table_v2 test=develop

* modify into PADDLE_ENFORCE_CUDA_SUCCESS test=develop

* Revert "modify into PADDLE_ENFORCE_CUDA_SUCCESS test=develop"

This reverts commit 8a960bfc61e51aa27c3c529df8fb90b93ebd19f9.

* move api into fluid.embedding test=develop

* fix example code test=develop

* move one_hot into fluid.one_hot

* modify api.spec test=develop

* fix loss shape test=develop

* support change shuffle thread num
* support change train thread num
* fix receive shuffle data of each channel
* data norm stop gradient
* add check thread_tensor type and root_tensor type when merge metric
* remove sleep in shuffle, add config
* add config of pslib client to client communication
* fix xbox str
* add data norm op testcase
* add flush in trainer finalize