compute and schedule updated for yolo reorg

Author: siju-samuel

include/tvm/relay/attrs/vision.h

@@ -53,9 +53,12 @@ struct NMSAttrs : public tvm::AttrsNode<NMSAttrs>{
         .describe("Suppress all detections regardless of class_id.");
       TVM_ATTR_FIELD(topk).set_default(-1)
         .describe("Keep maximum top k detections before nms, -1 for no limit.");
+  }
+};
+
 /*! \brief Attributes used in yolo reorg operators */
 struct YoloReorgAttrs : public tvm::AttrsNode<YoloReorgAttrs> {
-  IndexExpr stride;
+  Integer stride;
 
   TVM_DECLARE_ATTRS(YoloReorgAttrs, "relay.attrs.YoloReorgAttrs") {
     TVM_ATTR_FIELD(stride)

python/tvm/relay/op/vision/__init__.py

@@ -5,3 +5,4 @@
 from .multibox import *
 from .nms import *
 from .yolo import *
+from ._yolo import *

python/tvm/relay/op/vision/_yolo.py

@@ -0,0 +1,9 @@
+#pylint: disable=invalid-name, unused-argument
+"""Backend compiler related feature registration"""
+from __future__ import absolute_import
+from ..op import  register_schedule, register_pattern
+from ..op import schedule_injective, OpPattern
+
+# reorg
+register_pattern("vision.yolo_reorg", OpPattern.INJECTIVE)
+register_schedule("vision.yolo_reorg", schedule_injective)

src/relay/op/vision/yolo.cc

@@ -5,6 +5,7 @@
  */
 #include <tvm/relay/op.h>
 #include <tvm/relay/attrs/vision.h>
+#include <topi/vision/reorg.h>
 #include <vector>
 #include "../op_common.h"
 #include "../type_relations.h"
@@ -42,7 +43,7 @@ bool YoloReorgRel(const Array<Type>& types,
 }
 
 Expr MakeYoloReorg(Expr data,
-                   IndexExpr stride) {
+                   Integer stride) {
   auto attrs = make_node<YoloReorgAttrs>();
   attrs->stride = stride;
   static const Op& op = Op::Get("vision.yolo_reorg");
@@ -63,7 +64,15 @@ Its function is mostly shape transform.")doc" TVM_ADD_FILELINE)
 .set_num_inputs(1)
 .set_support_level(5)
 .set_attrs_type_key("relay.attrs.YoloReorgAttrs")
-.add_type_rel("YoloReorg", YoloReorgRel);
+.add_type_rel("YoloReorg", YoloReorgRel)
+.set_attr<FTVMCompute>("FTVMCompute", [](const Attrs& attrs,
+                                         const Array<Tensor>& inputs,
+                                         const Type& out_type,
+                                         const Target& target) {
+  const auto* params = attrs.as<YoloReorgAttrs>();
+  CHECK(params != nullptr);
+  return Array<Tensor>{ topi::vision::reorg(inputs[0], params->stride) };
+});
 
 }  // namespace relay
 }  // namespace tvm

tests/python/relay/test_op_level5.py

@@ -1,7 +1,10 @@
 """ Support level5 operator test cases.
 """
+import numpy as np
 import tvm
 from tvm import relay
+from tvm.relay.testing import ctx_list
+import topi.testing
 
 def test_resize_infer_type():
     n, c, h, w = tvm.var("n"), tvm.var("c"), tvm.var("h"), tvm.var("w")
@@ -70,22 +73,45 @@ def test_nms():
     zz = relay.ir_pass.infer_type(z)
     assert zz.checked_type == relay.ty.TensorType(
         (n, num_anchors, 6), "float32")
-def test_yolo_reorg():
+
+
+def test_yolo_reorg_infer_shape():
+    def verify_yolo_reorg(shape, stride, out_shape):
+        x = relay.var("x", relay.TensorType(shape, "float32"))
+        z = relay.vision.yolo_reorg(x, stride=stride)
+        zz = relay.ir_pass.infer_type(z)
+        assert "stride=" in z.astext()
+        assert zz.checked_type == relay.ty.TensorType(out_shape, "float32")
+
     n, c, h, w = tvm.var("n"), tvm.var("c"), tvm.var("h"), tvm.var("w")
-    x = relay.var("x", relay.TensorType((n, c, 20, 20), "float32"))
-    z = relay.vision.yolo_reorg(x, stride=10)
-    zz = relay.ir_pass.infer_type(z)
-    assert "stride=10" in z.astext()
-    assert zz.checked_type == relay.ty.TensorType((n, c*10*10, 2, 2), "float32")
+    verify_yolo_reorg((n, c, 20, 20), 10, (n, c*10*10, 2, 2))
+    verify_yolo_reorg((n, c, h, w), 2, (n, c*2*2, h/2, w/2))
 
-    x = relay.var("x", relay.TensorType((n, c, h, w), "float32"))
-    z = relay.vision.yolo_reorg(x, stride=2)
-    assert "stride=2" in z.astext()
-    zz = relay.ir_pass.infer_type(z)
-    assert zz.checked_type == relay.ty.TensorType((n, c*2*2, h/2, w/2), "float32")
+def test_yolo_reorg():
+    def verify_yolo_reorg(shape, stride):
+        x_data = np.random.uniform(low=-1, high=1, size=shape).astype("float32")
+        ref_res = topi.testing.reorg_python(x_data, stride)
+
+        x = relay.var("x", relay.TensorType(shape, "float32"))
+        z = relay.vision.yolo_reorg(x, stride=stride)
+        zz = relay.ir_pass.infer_type(z)
+        assert "stride=" in z.astext()
+        assert zz.checked_type == relay.ty.TensorType(ref_res.shape, "float32")
+
+        func = relay.Function([x], z)
+
+        for target, ctx in ctx_list():
+            for kind in ["graph", "debug"]:
+                intrp = relay.create_executor(kind, ctx=ctx, target=target)
+                op_res = intrp.evaluate(func)(x_data)
+                tvm.testing.assert_allclose(op_res.asnumpy(), ref_res, rtol=1e-5)
+
+    verify_yolo_reorg((1, 100, 20, 20), 10)
+    verify_yolo_reorg((1, 4, 6, 6), 2)
 
 if __name__ == "__main__":
     test_resize_infer_type()
     test_multibox_prior()
     test_nms()
+    test_yolo_reorg_infer_shape()
     test_yolo_reorg()