[TensorFlow] Fix limitation that depth_mult can only be 1 for Depthwi…

…seConv2dNative (apache#3676)

* [TensorFlow] Fix limitation that depth_mult can only be 1 for DepthwiseConv2dNative

* Improve code readability

python/tvm/relay/frontend/tensorflow.py

@@ -251,9 +251,6 @@ def _impl(inputs, attr, params):
             raise tvm.error.OpAttributeInvalid(msg.format(attr['data_format']))
 
         if opname == 'depthwise':
-            if depth_mult > 1:
-                raise tvm.error.OpNotImplemented('depth_mult > 1 of operator DepthwiseConv2dNative'
-                                                 ' is not supported.')
             attr['groups'] = attr['channels']
 
         # Fix padding

python/tvm/relay/op/nn/_nn.py

@@ -19,7 +19,7 @@
 from __future__ import absolute_import
 
 import topi
-from topi.util import get_const_int, get_const_tuple
+from topi.util import get_const_tuple
 from .. import op as reg
 from ..op import OpPattern, schedule_injective
 
@@ -144,19 +144,20 @@ def compute_conv2d(attrs, inputs, out_type, target):
     if dilation_h < 1 or dilation_w < 1:
         raise ValueError("dilation should be positive value")
 
+    def _get_out_depth():
+        weight_shape = get_const_tuple(inputs[1].shape)
+        if kernel_layout == "HWOI":
+            return weight_shape[2] * weight_shape[3]
+        return weight_shape[0] * weight_shape[1]
+
     if groups == 1:
         out = topi.nn.conv2d(
             inputs[0], inputs[1], strides, padding,
             dilation, layout, out_dtype)
-    elif layout == "NCHW" and \
-            get_const_int(inputs[1].shape[0]) == groups and \
-            get_const_int(inputs[1].shape[1]) == 1:
+    elif layout == "NCHW" and _get_out_depth() == groups:
         out = topi.nn.depthwise_conv2d_nchw(
             inputs[0], inputs[1], strides, padding, dilation, out_dtype)
-    elif layout == "NHWC" and \
-            kernel_layout == "HWOI" and\
-            get_const_int(inputs[1].shape[2]) == groups and \
-            get_const_int(inputs[1].shape[3]) == 1:
+    elif layout == "NHWC" and kernel_layout == "HWOI" and _get_out_depth() == groups:
         out = topi.nn.depthwise_conv2d_nhwc(
             inputs[0], inputs[1], strides, padding, dilation, out_dtype)
     elif layout in ['NCHW', 'NCHW4c']:

src/relay/op/nn/convolution.cc

@@ -23,6 +23,7 @@
  * \brief Convolution operators
  */
 #include <tvm/data_layout.h>
+#include <tvm/ir_pass.h>
 #include <tvm/relay/op.h>
 #include <tvm/relay/attrs/nn.h>
 #include <vector>
@@ -74,11 +75,23 @@ bool Conv2DRel(const Array<Type>& types,
   if (param->kernel_size.defined() && param->channels.defined()) {
     CHECK_EQ(param->kernel_size.size(), 2);
     CHECK_EQ(param->dilation.size(), 2);
-    Array<IndexExpr> wshape(
-       {param->channels,
-         dshape_nchw[1] / param->groups,
-         param->kernel_size[0],
-         param->kernel_size[1]});
+    Array<IndexExpr> wshape;
+
+    if (tvm::ir::Equal(param->channels, param->groups)) {
+      // infer weight's shape for depthwise convolution
+      wshape = {
+         {dshape_nchw[1],
+          param->groups / dshape_nchw[1],
+          param->kernel_size[0],
+          param->kernel_size[1]}};
+    } else {
+      wshape = {
+         {param->channels,
+          dshape_nchw[1] / param->groups,
+          param->kernel_size[0],
+          param->kernel_size[1]}};
+    }
+
     wshape = trans_kernel_layout.BackwardShape(wshape);
     channels = param->channels;
     dilated_ksize_y = 1 + (param->kernel_size[0] - 1) * param->dilation[0];

tests/python/frontend/tensorflow/test_forward.py

@@ -275,6 +275,7 @@ def test_forward_convolution():
         _test_convolution('depthwise', [4, 19, 17, 17], [3, 3, 19, 1], [1, 1], [2, 2], 'VALID', 'NCHW')
         _test_convolution('depthwise', [4, 124, 17, 17], [1, 1, 124, 1], [1, 1], [1, 1], 'SAME', 'NCHW')
         _test_convolution('depthwise', [4, 12, 17, 17], [3, 3, 12, 1], [1, 1], [2, 2], 'VALID', 'NCHW')
+        _test_convolution('depthwise', [4, 12, 17, 17], [3, 3, 12, 2], [1, 1], [2, 2], 'VALID', 'NCHW')
 
     _test_convolution('conv', [4, 8, 8, 176], [1, 1, 176, 32], [1, 1], [1, 1], 'SAME', 'NHWC')
     _test_convolution('conv', [4, 17, 17, 19], [3, 3, 19, 19], [1, 1], [2, 2], 'VALID', 'NHWC')
@@ -284,6 +285,7 @@ def test_forward_convolution():
     _test_convolution('depthwise', [4, 17, 17, 19], [3, 3, 19, 1], [1, 1], [2, 2], 'VALID', 'NHWC')
     _test_convolution('depthwise', [4, 17, 17, 124], [1, 1, 124, 1], [1, 1], [1, 1], 'SAME', 'NHWC')
     _test_convolution('depthwise', [4, 17, 17, 12], [3, 3, 12, 1], [1, 1], [2, 2], 'VALID', 'NHWC')
+    _test_convolution('depthwise', [4, 17, 17, 12], [3, 3, 12, 2], [1, 1], [2, 2], 'VALID', 'NHWC')
 
 #######################################################################
 # BiasAdd