* Fixed issue #3069 by adding in_channels

* Registerd group_conv2d_nchw as topi compute

* Improved by checking tag value

* Removed group_conv2d_nchw topi registration

* Added test for relay group_conv2d_nchw

* Added assertions to forbid small group size

* Removed hard-coded oc_block_factor

* Added explanatory comments to group_conv2d_nchw_cuda

* Updated group_conv2d_nchw_cuda schedule

Removed 'direct' CUDA tests

* Reverted an accidental change in a conv2d test

* Fixed indentation problems

* Fixed a mis-commented line

* Reverted change in group_conv2d_nchw tag

* Removed commented int8 group_conv2d test

* Fixed group size assertions in group_conv2d_nchw_cuda

1) fixed a functional bug in loop partitioning algorithm that is exposed when double splitting with indivisible factors 2) added a testcase (#2956)

[TVM][ARITH] Teach BoundDeduce to handle the case in which target var can appear in rhs of expression (#2795)

* target variable can now appear in either lhs or rhs of the expression to be analyzed

* removed extra spaces

* Quantize dense layers

* Add out_dtype arggument to dense; Add dense_int8 on CUDA

* Add topi unittest of dense int8

* Fix relay

* Fix topi integration

* Fix quantization

* Update dense_rewrite

* Triger CI

* Change qconfig quantize_dense to quantize_op

* Fix

* Remove quantize_op from qconfig

* [Relay][TensorFlow] Remove 'input_0d_mismatch' special handling

* Add more tests.

* Cover the case that strided_slice outputs a scalar

* Enhance upsample operator to adapt onnx opset version 9 for nnvm compiler

* Add upsample test case for newer opset in nnvm

* re-trigger the CI

* [Frontend][TF] Fix Placeholder issue

* Add test cases

* [Relay][Frontend] TF Round Sign Pow Exp Reverse

* fix ci

* fix comments

* Update pretty_printer.cc

* Update pretty_printer.cc

* Implement nn.bias_add compute in C++

* Address comments

* Remove unnecessary check

* save

* fix

* Update fuse_ops.cc