Add autocast for nms, roi_align on CPU (#8049)

Co-authored-by: Nicolas Hug <[email protected]>
Co-authored-by: Nicolas Hug <[email protected]>

setup.py

@@ -129,8 +129,10 @@ def get_extensions():
     this_dir = os.path.dirname(os.path.abspath(__file__))
     extensions_dir = os.path.join(this_dir, "torchvision", "csrc")
 
-    main_file = glob.glob(os.path.join(extensions_dir, "*.cpp")) + glob.glob(
-        os.path.join(extensions_dir, "ops", "*.cpp")
+    main_file = (
+        glob.glob(os.path.join(extensions_dir, "*.cpp"))
+        + glob.glob(os.path.join(extensions_dir, "ops", "*.cpp"))
+        + glob.glob(os.path.join(extensions_dir, "ops", "autocast", "*.cpp"))
     )
     source_cpu = (
         glob.glob(os.path.join(extensions_dir, "ops", "autograd", "*.cpp"))
@@ -184,8 +186,6 @@ def get_extensions():
     else:
         source_cuda = glob.glob(os.path.join(extensions_dir, "ops", "cuda", "*.cu"))
 
-    source_cuda += glob.glob(os.path.join(extensions_dir, "ops", "autocast", "*.cpp"))
-
     sources = main_file + source_cpu
     extension = CppExtension
 

test/test_ops.py

@@ -131,6 +131,8 @@ def test_forward(self, device, contiguous, x_dtype, rois_dtype=None, determinist
                 tol = 5e-3
             else:
                 tol = 4e-3
+        elif x_dtype == torch.bfloat16:
+            tol = 5e-3
 
         pool_size = 5
         # n_channels % (pool_size ** 2) == 0 required for PS operations.
@@ -504,6 +506,21 @@ def test_autocast(self, aligned, deterministic, x_dtype, rois_dtype):
                 rois_dtype=rois_dtype,
             )
 
+    @pytest.mark.parametrize("aligned", (True, False))
+    @pytest.mark.parametrize("deterministic", (True, False))
+    @pytest.mark.parametrize("x_dtype", (torch.float, torch.bfloat16))
+    @pytest.mark.parametrize("rois_dtype", (torch.float, torch.bfloat16))
+    def test_autocast_cpu(self, aligned, deterministic, x_dtype, rois_dtype):
+        with torch.cpu.amp.autocast():
+            self.test_forward(
+                torch.device("cpu"),
+                contiguous=False,
+                deterministic=deterministic,
+                aligned=aligned,
+                x_dtype=x_dtype,
+                rois_dtype=rois_dtype,
+            )
+
     @pytest.mark.parametrize("seed", range(10))
     @pytest.mark.parametrize("device", cpu_and_cuda_and_mps())
     @pytest.mark.parametrize("contiguous", (True, False))
@@ -808,6 +825,15 @@ def test_autocast(self, iou, dtype):
         with torch.cuda.amp.autocast():
             self.test_nms_gpu(iou=iou, dtype=dtype, device="cuda")
 
+    @pytest.mark.parametrize("iou", (0.2, 0.5, 0.8))
+    @pytest.mark.parametrize("dtype", (torch.float, torch.bfloat16))
+    def test_autocast_cpu(self, iou, dtype):
+        boxes, scores = self._create_tensors_with_iou(1000, iou)
+        with torch.cpu.amp.autocast():
+            keep_ref_float = ops.nms(boxes.to(dtype).float(), scores.to(dtype).float(), iou)
+            keep_dtype = ops.nms(boxes.to(dtype), scores.to(dtype), iou)
+        torch.testing.assert_close(keep_ref_float, keep_dtype)
+
     @pytest.mark.parametrize(
         "device",
         (

torchvision/csrc/ops/autocast/nms_kernel.cpp

@@ -9,21 +9,33 @@ namespace ops {
 
 namespace {
 
+template <c10::DispatchKey autocast_key, c10::DeviceType device_type>
 at::Tensor nms_autocast(
     const at::Tensor& dets,
     const at::Tensor& scores,
     double iou_threshold) {
-  c10::impl::ExcludeDispatchKeyGuard no_autocast(c10::DispatchKey::Autocast);
+  c10::impl::ExcludeDispatchKeyGuard no_autocast(autocast_key);
+
   return nms(
-      at::autocast::cached_cast(at::kFloat, dets),
-      at::autocast::cached_cast(at::kFloat, scores),
+      at::autocast::cached_cast(at::kFloat, dets, device_type),
+      at::autocast::cached_cast(at::kFloat, scores, device_type),
       iou_threshold);
 }
 
 } // namespace
 
 TORCH_LIBRARY_IMPL(torchvision, Autocast, m) {
-  m.impl(TORCH_SELECTIVE_NAME("torchvision::nms"), TORCH_FN(nms_autocast));
+  m.impl(
+      TORCH_SELECTIVE_NAME("torchvision::nms"),
+      TORCH_FN(
+          (nms_autocast<c10::DispatchKey::Autocast, c10::DeviceType::CUDA>)));
+}
+
+TORCH_LIBRARY_IMPL(torchvision, AutocastCPU, m) {
+  m.impl(
+      TORCH_SELECTIVE_NAME("torchvision::nms"),
+      TORCH_FN(
+          (nms_autocast<c10::DispatchKey::AutocastCPU, c10::DeviceType::CPU>)));
 }
 
 } // namespace ops

torchvision/csrc/ops/autocast/roi_align_kernel.cpp

@@ -9,6 +9,7 @@ namespace ops {
 
 namespace {
 
+template <c10::DispatchKey autocast_key, c10::DeviceType device_type>
 at::Tensor roi_align_autocast(
     const at::Tensor& input,
     const at::Tensor& rois,
@@ -17,10 +18,10 @@ at::Tensor roi_align_autocast(
     int64_t pooled_width,
     int64_t sampling_ratio,
     bool aligned) {
-  c10::impl::ExcludeDispatchKeyGuard no_autocast(c10::DispatchKey::Autocast);
+  c10::impl::ExcludeDispatchKeyGuard no_autocast(autocast_key);
   return roi_align(
-             at::autocast::cached_cast(at::kFloat, input),
-             at::autocast::cached_cast(at::kFloat, rois),
+             at::autocast::cached_cast(at::kFloat, input, device_type),
+             at::autocast::cached_cast(at::kFloat, rois, device_type),
              spatial_scale,
              pooled_height,
              pooled_width,
@@ -34,7 +35,17 @@ at::Tensor roi_align_autocast(
 TORCH_LIBRARY_IMPL(torchvision, Autocast, m) {
   m.impl(
       TORCH_SELECTIVE_NAME("torchvision::roi_align"),
-      TORCH_FN(roi_align_autocast));
+      TORCH_FN((roi_align_autocast<
+                c10::DispatchKey::Autocast,
+                c10::DeviceType::CUDA>)));
+}
+
+TORCH_LIBRARY_IMPL(torchvision, AutocastCPU, m) {
+  m.impl(
+      TORCH_SELECTIVE_NAME("torchvision::roi_align"),
+      TORCH_FN((roi_align_autocast<
+                c10::DispatchKey::AutocastCPU,
+                c10::DeviceType::CPU>)));
 }
 
 } // namespace ops