test_ops.py

from common_utils import needs_cuda, cpu_and_gpu
from _assert_utils import assert_equal
import math
from abc import ABC, abstractmethod
import pytest

import numpy as np

import torch
from functools import lru_cache
from torch import Tensor
from torch.autograd import gradcheck
from torch.nn.modules.utils import _pair
from torchvision import ops
from typing import Tuple


class RoIOpTester(ABC):
    dtype = torch.float64

    @pytest.mark.parametrize('device', cpu_and_gpu())
    @pytest.mark.parametrize('contiguous', (True, False))
    def test_forward(self, device, contiguous, x_dtype=None, rois_dtype=None, **kwargs):
        x_dtype = self.dtype if x_dtype is None else x_dtype
        rois_dtype = self.dtype if rois_dtype is None else rois_dtype
        pool_size = 5
        # n_channels % (pool_size ** 2) == 0 required for PS opeartions.
        n_channels = 2 * (pool_size ** 2)
        x = torch.rand(2, n_channels, 10, 10, dtype=x_dtype, device=device)
        if not contiguous:
            x = x.permute(0, 1, 3, 2)
        rois = torch.tensor([[0, 0, 0, 9, 9],  # format is (xyxy)
                             [0, 0, 5, 4, 9],
                             [0, 5, 5, 9, 9],
                             [1, 0, 0, 9, 9]],
                            dtype=rois_dtype, device=device)

        pool_h, pool_w = pool_size, pool_size
        y = self.fn(x, rois, pool_h, pool_w, spatial_scale=1, sampling_ratio=-1, **kwargs)
        # the following should be true whether we're running an autocast test or not.
        assert y.dtype == x.dtype
        gt_y = self.expected_fn(x, rois, pool_h, pool_w, spatial_scale=1,
                                sampling_ratio=-1, device=device, dtype=self.dtype, **kwargs)

        tol = 1e-3 if (x_dtype is torch.half or rois_dtype is torch.half) else 1e-5
        torch.testing.assert_close(gt_y.to(y), y, rtol=tol, atol=tol)

    @pytest.mark.parametrize('device', cpu_and_gpu())
    @pytest.mark.parametrize('contiguous', (True, False))
    def test_backward(self, device, contiguous):
        pool_size = 2
        x = torch.rand(1, 2 * (pool_size ** 2), 5, 5, dtype=self.dtype, device=device, requires_grad=True)
        if not contiguous:
            x = x.permute(0, 1, 3, 2)
        rois = torch.tensor([[0, 0, 0, 4, 4],  # format is (xyxy)
                             [0, 0, 2, 3, 4],
                             [0, 2, 2, 4, 4]],
                            dtype=self.dtype, device=device)

        def func(z):
            return self.fn(z, rois, pool_size, pool_size, spatial_scale=1, sampling_ratio=1)

        script_func = self.get_script_fn(rois, pool_size)

        gradcheck(func, (x,))
        gradcheck(script_func, (x,))

    @needs_cuda
    @pytest.mark.parametrize('x_dtype', (torch.float, torch.half))
    @pytest.mark.parametrize('rois_dtype', (torch.float, torch.half))
    def test_autocast(self, x_dtype, rois_dtype):
        with torch.cuda.amp.autocast():
            self.test_forward(torch.device("cuda"), contiguous=False, x_dtype=x_dtype, rois_dtype=rois_dtype)

    def _helper_boxes_shape(self, func):
        # test boxes as Tensor[N, 5]
        with pytest.raises(AssertionError):
            a = torch.linspace(1, 8 * 8, 8 * 8).reshape(1, 1, 8, 8)
            boxes = torch.tensor([[0, 0, 3, 3]], dtype=a.dtype)
            func(a, boxes, output_size=(2, 2))

        # test boxes as List[Tensor[N, 4]]
        with pytest.raises(AssertionError):
            a = torch.linspace(1, 8 * 8, 8 * 8).reshape(1, 1, 8, 8)
            boxes = torch.tensor([[0, 0, 3]], dtype=a.dtype)
            ops.roi_pool(a, [boxes], output_size=(2, 2))

    @abstractmethod
    def fn(*args, **kwargs):
        pass

    @abstractmethod
    def get_script_fn(*args, **kwargs):
        pass

    @abstractmethod
    def expected_fn(*args, **kwargs):
        pass


class TestRoiPool(RoIOpTester):
    def fn(self, x, rois, pool_h, pool_w, spatial_scale=1, sampling_ratio=-1, **kwargs):
        return ops.RoIPool((pool_h, pool_w), spatial_scale)(x, rois)

    def get_script_fn(self, rois, pool_size):
        scriped = torch.jit.script(ops.roi_pool)
        return lambda x: scriped(x, rois, pool_size)

    def expected_fn(self, x, rois, pool_h, pool_w, spatial_scale=1, sampling_ratio=-1,
                    device=None, dtype=torch.float64):
        if device is None:
            device = torch.device("cpu")

        n_channels = x.size(1)
        y = torch.zeros(rois.size(0), n_channels, pool_h, pool_w, dtype=dtype, device=device)

        def get_slice(k, block):
            return slice(int(np.floor(k * block)), int(np.ceil((k + 1) * block)))

        for roi_idx, roi in enumerate(rois):
            batch_idx = int(roi[0])
            j_begin, i_begin, j_end, i_end = (int(round(x.item() * spatial_scale)) for x in roi[1:])
            roi_x = x[batch_idx, :, i_begin:i_end + 1, j_begin:j_end + 1]

            roi_h, roi_w = roi_x.shape[-2:]
            bin_h = roi_h / pool_h
            bin_w = roi_w / pool_w

            for i in range(0, pool_h):
                for j in range(0, pool_w):
                    bin_x = roi_x[:, get_slice(i, bin_h), get_slice(j, bin_w)]
                    if bin_x.numel() > 0:
                        y[roi_idx, :, i, j] = bin_x.reshape(n_channels, -1).max(dim=1)[0]
        return y

    def test_boxes_shape(self):
        self._helper_boxes_shape(ops.roi_pool)


class TestPSRoIPool(RoIOpTester):
    def fn(self, x, rois, pool_h, pool_w, spatial_scale=1, sampling_ratio=-1, **kwargs):
        return ops.PSRoIPool((pool_h, pool_w), 1)(x, rois)

    def get_script_fn(self, rois, pool_size):
        scriped = torch.jit.script(ops.ps_roi_pool)
        return lambda x: scriped(x, rois, pool_size)

    def expected_fn(self, x, rois, pool_h, pool_w, spatial_scale=1, sampling_ratio=-1,
                    device=None, dtype=torch.float64):
        if device is None:
            device = torch.device("cpu")
        n_input_channels = x.size(1)
        assert n_input_channels % (pool_h * pool_w) == 0, "input channels must be divisible by ph * pw"
        n_output_channels = int(n_input_channels / (pool_h * pool_w))
        y = torch.zeros(rois.size(0), n_output_channels, pool_h, pool_w, dtype=dtype, device=device)

        def get_slice(k, block):
            return slice(int(np.floor(k * block)), int(np.ceil((k + 1) * block)))

        for roi_idx, roi in enumerate(rois):
            batch_idx = int(roi[0])
            j_begin, i_begin, j_end, i_end = (int(round(x.item() * spatial_scale)) for x in roi[1:])
            roi_x = x[batch_idx, :, i_begin:i_end + 1, j_begin:j_end + 1]

            roi_height = max(i_end - i_begin, 1)
            roi_width = max(j_end - j_begin, 1)
            bin_h, bin_w = roi_height / float(pool_h), roi_width / float(pool_w)

            for i in range(0, pool_h):
                for j in range(0, pool_w):
                    bin_x = roi_x[:, get_slice(i, bin_h), get_slice(j, bin_w)]
                    if bin_x.numel() > 0:
                        area = bin_x.size(-2) * bin_x.size(-1)
                        for c_out in range(0, n_output_channels):
                            c_in = c_out * (pool_h * pool_w) + pool_w * i + j
                            t = torch.sum(bin_x[c_in, :, :])
                            y[roi_idx, c_out, i, j] = t / area
        return y

    def test_boxes_shape(self):
        self._helper_boxes_shape(ops.ps_roi_pool)


def bilinear_interpolate(data, y, x, snap_border=False):
    height, width = data.shape

    if snap_border:
        if -1 < y <= 0:
            y = 0
        elif height - 1 <= y < height:
            y = height - 1

        if -1 < x <= 0:
            x = 0
        elif width - 1 <= x < width:
            x = width - 1

    y_low = int(math.floor(y))
    x_low = int(math.floor(x))
    y_high = y_low + 1
    x_high = x_low + 1

    wy_h = y - y_low
    wx_h = x - x_low
    wy_l = 1 - wy_h
    wx_l = 1 - wx_h

    val = 0
    for wx, xp in zip((wx_l, wx_h), (x_low, x_high)):
        for wy, yp in zip((wy_l, wy_h), (y_low, y_high)):
            if 0 <= yp < height and 0 <= xp < width:
                val += wx * wy * data[yp, xp]
    return val


class TestRoIAlign(RoIOpTester):
    def fn(self, x, rois, pool_h, pool_w, spatial_scale=1, sampling_ratio=-1, aligned=False, **kwargs):
        return ops.RoIAlign((pool_h, pool_w), spatial_scale=spatial_scale,
                            sampling_ratio=sampling_ratio, aligned=aligned)(x, rois)

    def get_script_fn(self, rois, pool_size):
        scriped = torch.jit.script(ops.roi_align)
        return lambda x: scriped(x, rois, pool_size)

    def expected_fn(self, in_data, rois, pool_h, pool_w, spatial_scale=1, sampling_ratio=-1, aligned=False,
                    device=None, dtype=torch.float64):
        if device is None:
            device = torch.device("cpu")
        n_channels = in_data.size(1)
        out_data = torch.zeros(rois.size(0), n_channels, pool_h, pool_w, dtype=dtype, device=device)

        offset = 0.5 if aligned else 0.

        for r, roi in enumerate(rois):
            batch_idx = int(roi[0])
            j_begin, i_begin, j_end, i_end = (x.item() * spatial_scale - offset for x in roi[1:])

            roi_h = i_end - i_begin
            roi_w = j_end - j_begin
            bin_h = roi_h / pool_h
            bin_w = roi_w / pool_w

            for i in range(0, pool_h):
                start_h = i_begin + i * bin_h
                grid_h = sampling_ratio if sampling_ratio > 0 else int(np.ceil(bin_h))
                for j in range(0, pool_w):
                    start_w = j_begin + j * bin_w
                    grid_w = sampling_ratio if sampling_ratio > 0 else int(np.ceil(bin_w))

                    for channel in range(0, n_channels):

                        val = 0
                        for iy in range(0, grid_h):
                            y = start_h + (iy + 0.5) * bin_h / grid_h
                            for ix in range(0, grid_w):
                                x = start_w + (ix + 0.5) * bin_w / grid_w
                                val += bilinear_interpolate(in_data[batch_idx, channel, :, :], y, x, snap_border=True)
                        val /= grid_h * grid_w

                        out_data[r, channel, i, j] = val
        return out_data

    def test_boxes_shape(self):
        self._helper_boxes_shape(ops.roi_align)

    @pytest.mark.parametrize('aligned', (True, False))
    @pytest.mark.parametrize('device', cpu_and_gpu())
    @pytest.mark.parametrize('contiguous', (True, False))
    def test_forward(self, device, contiguous, aligned, x_dtype=None, rois_dtype=None):
        super().test_forward(device=device, contiguous=contiguous, x_dtype=x_dtype, rois_dtype=rois_dtype,
                             aligned=aligned)

    @needs_cuda
    @pytest.mark.parametrize('aligned', (True, False))
    @pytest.mark.parametrize('x_dtype', (torch.float, torch.half))
    @pytest.mark.parametrize('rois_dtype', (torch.float, torch.half))
    def test_autocast(self, aligned, x_dtype, rois_dtype):
        with torch.cuda.amp.autocast():
            self.test_forward(torch.device("cuda"), contiguous=False, aligned=aligned, x_dtype=x_dtype,
                              rois_dtype=rois_dtype)

    def _make_rois(self, img_size, num_imgs, dtype, num_rois=1000):
        rois = torch.randint(0, img_size // 2, size=(num_rois, 5)).to(dtype)
        rois[:, 0] = torch.randint(0, num_imgs, size=(num_rois,))  # set batch index
        rois[:, 3:] += rois[:, 1:3]  # make sure boxes aren't degenerate
        return rois

    @pytest.mark.parametrize('aligned', (True, False))
    @pytest.mark.parametrize('scale, zero_point', ((1, 0), (2, 10), (0.1, 50)))
    @pytest.mark.parametrize('qdtype', (torch.qint8, torch.quint8, torch.qint32))
    def test_qroialign(self, aligned, scale, zero_point, qdtype):
        """Make sure quantized version of RoIAlign is close to float version"""
        pool_size = 5
        img_size = 10
        n_channels = 2
        num_imgs = 1
        dtype = torch.float

        x = torch.randint(50, 100, size=(num_imgs, n_channels, img_size, img_size)).to(dtype)
        qx = torch.quantize_per_tensor(x, scale=scale, zero_point=zero_point, dtype=qdtype)

        rois = self._make_rois(img_size, num_imgs, dtype)
        qrois = torch.quantize_per_tensor(rois, scale=scale, zero_point=zero_point, dtype=qdtype)

        x, rois = qx.dequantize(), qrois.dequantize()  # we want to pass the same inputs

        y = ops.roi_align(
            x,
            rois,
            output_size=pool_size,
            spatial_scale=1,
            sampling_ratio=-1,
            aligned=aligned,
        )
        qy = ops.roi_align(
            qx,
            qrois,
            output_size=pool_size,
            spatial_scale=1,
            sampling_ratio=-1,
            aligned=aligned,
        )

        # The output qy is itself a quantized tensor and there might have been a loss of info when it was
        # quantized. For a fair comparison we need to quantize y as well
        quantized_float_y = torch.quantize_per_tensor(y, scale=scale, zero_point=zero_point, dtype=qdtype)

        try:
            # Ideally, we would assert this, which passes with (scale, zero) == (1, 0)
            assert (qy == quantized_float_y).all()
        except AssertionError:
            # But because the computation aren't exactly the same between the 2 RoIAlign procedures, some
            # rounding error may lead to a difference of 2 in the output.
            # For example with (scale, zero) = (2, 10), 45.00000... will be quantized to 44
            # but 45.00000001 will be rounded to 46. We make sure below that:
            # - such discrepancies between qy and quantized_float_y are very rare (less then 5%)
            # - any difference between qy and quantized_float_y is == scale
            diff_idx = torch.where(qy != quantized_float_y)
            num_diff = diff_idx[0].numel()
            assert num_diff / qy.numel() < .05

            abs_diff = torch.abs(qy[diff_idx].dequantize() - quantized_float_y[diff_idx].dequantize())
            t_scale = torch.full_like(abs_diff, fill_value=scale)
            torch.testing.assert_close(abs_diff, t_scale, rtol=1e-5, atol=1e-5)

    def test_qroi_align_multiple_images(self):
        dtype = torch.float
        x = torch.randint(50, 100, size=(2, 3, 10, 10)).to(dtype)
        qx = torch.quantize_per_tensor(x, scale=1, zero_point=0, dtype=torch.qint8)
        rois = self._make_rois(img_size=10, num_imgs=2, dtype=dtype, num_rois=10)
        qrois = torch.quantize_per_tensor(rois, scale=1, zero_point=0, dtype=torch.qint8)
        with pytest.raises(RuntimeError, match="Only one image per batch is allowed"):
            ops.roi_align(qx, qrois, output_size=5)


class TestPSRoIAlign(RoIOpTester):
    def fn(self, x, rois, pool_h, pool_w, spatial_scale=1, sampling_ratio=-1, **kwargs):
        return ops.PSRoIAlign((pool_h, pool_w), spatial_scale=spatial_scale,
                              sampling_ratio=sampling_ratio)(x, rois)

    def get_script_fn(self, rois, pool_size):
        scriped = torch.jit.script(ops.ps_roi_align)
        return lambda x: scriped(x, rois, pool_size)

    def expected_fn(self, in_data, rois, pool_h, pool_w, device, spatial_scale=1,
                    sampling_ratio=-1, dtype=torch.float64):
        if device is None:
            device = torch.device("cpu")
        n_input_channels = in_data.size(1)
        assert n_input_channels % (pool_h * pool_w) == 0, "input channels must be divisible by ph * pw"
        n_output_channels = int(n_input_channels / (pool_h * pool_w))
        out_data = torch.zeros(rois.size(0), n_output_channels, pool_h, pool_w, dtype=dtype, device=device)

        for r, roi in enumerate(rois):
            batch_idx = int(roi[0])
            j_begin, i_begin, j_end, i_end = (x.item() * spatial_scale - 0.5 for x in roi[1:])

            roi_h = i_end - i_begin
            roi_w = j_end - j_begin
            bin_h = roi_h / pool_h
            bin_w = roi_w / pool_w

            for i in range(0, pool_h):
                start_h = i_begin + i * bin_h
                grid_h = sampling_ratio if sampling_ratio > 0 else int(np.ceil(bin_h))
                for j in range(0, pool_w):
                    start_w = j_begin + j * bin_w
                    grid_w = sampling_ratio if sampling_ratio > 0 else int(np.ceil(bin_w))
                    for c_out in range(0, n_output_channels):
                        c_in = c_out * (pool_h * pool_w) + pool_w * i + j

                        val = 0
                        for iy in range(0, grid_h):
                            y = start_h + (iy + 0.5) * bin_h / grid_h
                            for ix in range(0, grid_w):
                                x = start_w + (ix + 0.5) * bin_w / grid_w
                                val += bilinear_interpolate(in_data[batch_idx, c_in, :, :], y, x, snap_border=True)
                        val /= grid_h * grid_w

                        out_data[r, c_out, i, j] = val
        return out_data

    def test_boxes_shape(self):
        self._helper_boxes_shape(ops.ps_roi_align)


class TestMultiScaleRoIAlign:
    def test_msroialign_repr(self):
        fmap_names = ['0']
        output_size = (7, 7)
        sampling_ratio = 2
        # Pass mock feature map names
        t = ops.poolers.MultiScaleRoIAlign(fmap_names, output_size, sampling_ratio)

        # Check integrity of object __repr__ attribute
        expected_string = (f"MultiScaleRoIAlign(featmap_names={fmap_names}, output_size={output_size}, "
                           f"sampling_ratio={sampling_ratio})")
        assert repr(t) == expected_string


class TestNMS:
    def _reference_nms(self, boxes, scores, iou_threshold):
        """
        Args:
            box_scores (N, 5): boxes in corner-form and probabilities.
            iou_threshold: intersection over union threshold.
        Returns:
             picked: a list of indexes of the kept boxes
        """
        picked = []
        _, indexes = scores.sort(descending=True)
        while len(indexes) > 0:
            current = indexes[0]
            picked.append(current.item())
            if len(indexes) == 1:
                break
            current_box = boxes[current, :]
            indexes = indexes[1:]
            rest_boxes = boxes[indexes, :]
            iou = ops.box_iou(rest_boxes, current_box.unsqueeze(0)).squeeze(1)
            indexes = indexes[iou <= iou_threshold]

        return torch.as_tensor(picked)

    def _create_tensors_with_iou(self, N, iou_thresh):
        # force last box to have a pre-defined iou with the first box
        # let b0 be [x0, y0, x1, y1], and b1 be [x0, y0, x1 + d, y1],
        # then, in order to satisfy ops.iou(b0, b1) == iou_thresh,
        # we need to have d = (x1 - x0) * (1 - iou_thresh) / iou_thresh
        # Adjust the threshold upward a bit with the intent of creating
        # at least one box that exceeds (barely) the threshold and so
        # should be suppressed.
        boxes = torch.rand(N, 4) * 100
        boxes[:, 2:] += boxes[:, :2]
        boxes[-1, :] = boxes[0, :]
        x0, y0, x1, y1 = boxes[-1].tolist()
        iou_thresh += 1e-5
        boxes[-1, 2] += (x1 - x0) * (1 - iou_thresh) / iou_thresh
        scores = torch.rand(N)
        return boxes, scores

    @pytest.mark.parametrize("iou", (.2, .5, .8))
    def test_nms_ref(self, iou):
        err_msg = 'NMS incompatible between CPU and reference implementation for IoU={}'
        boxes, scores = self._create_tensors_with_iou(1000, iou)
        keep_ref = self._reference_nms(boxes, scores, iou)
        keep = ops.nms(boxes, scores, iou)
        assert torch.allclose(keep, keep_ref), err_msg.format(iou)

    def test_nms_input_errors(self):
        with pytest.raises(RuntimeError):
            ops.nms(torch.rand(4), torch.rand(3), 0.5)
        with pytest.raises(RuntimeError):
            ops.nms(torch.rand(3, 5), torch.rand(3), 0.5)
        with pytest.raises(RuntimeError):
            ops.nms(torch.rand(3, 4), torch.rand(3, 2), 0.5)
        with pytest.raises(RuntimeError):
            ops.nms(torch.rand(3, 4), torch.rand(4), 0.5)

    @pytest.mark.parametrize("iou", (.2, .5, .8))
    @pytest.mark.parametrize("scale, zero_point", ((1, 0), (2, 50), (3, 10)))
    def test_qnms(self, iou, scale, zero_point):
        # Note: we compare qnms vs nms instead of qnms vs reference implementation.
        # This is because with the int convertion, the trick used in _create_tensors_with_iou
        # doesn't really work (in fact, nms vs reference implem will also fail with ints)
        err_msg = 'NMS and QNMS give different results for IoU={}'
        boxes, scores = self._create_tensors_with_iou(1000, iou)
        scores *= 100  # otherwise most scores would be 0 or 1 after int convertion

        qboxes = torch.quantize_per_tensor(boxes, scale=scale, zero_point=zero_point, dtype=torch.quint8)
        qscores = torch.quantize_per_tensor(scores, scale=scale, zero_point=zero_point, dtype=torch.quint8)

        boxes = qboxes.dequantize()
        scores = qscores.dequantize()

        keep = ops.nms(boxes, scores, iou)
        qkeep = ops.nms(qboxes, qscores, iou)

        assert torch.allclose(qkeep, keep), err_msg.format(iou)

    @needs_cuda
    @pytest.mark.parametrize("iou", (.2, .5, .8))
    def test_nms_cuda(self, iou, dtype=torch.float64):
        tol = 1e-3 if dtype is torch.half else 1e-5
        err_msg = 'NMS incompatible between CPU and CUDA for IoU={}'

        boxes, scores = self._create_tensors_with_iou(1000, iou)
        r_cpu = ops.nms(boxes, scores, iou)
        r_cuda = ops.nms(boxes.cuda(), scores.cuda(), iou)

        is_eq = torch.allclose(r_cpu, r_cuda.cpu())
        if not is_eq:
            # if the indices are not the same, ensure that it's because the scores
            # are duplicate
            is_eq = torch.allclose(scores[r_cpu], scores[r_cuda.cpu()], rtol=tol, atol=tol)
        assert is_eq, err_msg.format(iou)

    @needs_cuda
    @pytest.mark.parametrize("iou", (.2, .5, .8))
    @pytest.mark.parametrize("dtype", (torch.float, torch.half))
    def test_autocast(self, iou, dtype):
        with torch.cuda.amp.autocast():
            self.test_nms_cuda(iou=iou, dtype=dtype)

    @needs_cuda
    def test_nms_cuda_float16(self):
        boxes = torch.tensor([[285.3538, 185.5758, 1193.5110, 851.4551],
                              [285.1472, 188.7374, 1192.4984, 851.0669],
                              [279.2440, 197.9812, 1189.4746, 849.2019]]).cuda()
        scores = torch.tensor([0.6370, 0.7569, 0.3966]).cuda()

        iou_thres = 0.2
        keep32 = ops.nms(boxes, scores, iou_thres)
        keep16 = ops.nms(boxes.to(torch.float16), scores.to(torch.float16), iou_thres)
        assert_equal(keep32, keep16)

    def test_batched_nms_implementations(self):
        """Make sure that both implementations of batched_nms yield identical results"""

        num_boxes = 1000
        iou_threshold = .9

        boxes = torch.cat((torch.rand(num_boxes, 2), torch.rand(num_boxes, 2) + 10), dim=1)
        assert max(boxes[:, 0]) < min(boxes[:, 2])  # x1 < x2
        assert max(boxes[:, 1]) < min(boxes[:, 3])  # y1 < y2

        scores = torch.rand(num_boxes)
        idxs = torch.randint(0, 4, size=(num_boxes,))
        keep_vanilla = ops.boxes._batched_nms_vanilla(boxes, scores, idxs, iou_threshold)
        keep_trick = ops.boxes._batched_nms_coordinate_trick(boxes, scores, idxs, iou_threshold)

        torch.testing.assert_close(
            keep_vanilla, keep_trick, msg="The vanilla and the trick implementation yield different nms outputs."
        )

        # Also make sure an empty tensor is returned if boxes is empty
        empty = torch.empty((0,), dtype=torch.int64)
        torch.testing.assert_close(empty, ops.batched_nms(empty, None, None, None))


class TestDeformConv:
    dtype = torch.float64

    def expected_fn(self, x, weight, offset, mask, bias, stride=1, padding=0, dilation=1):
        stride_h, stride_w = _pair(stride)
        pad_h, pad_w = _pair(padding)
        dil_h, dil_w = _pair(dilation)
        weight_h, weight_w = weight.shape[-2:]

        n_batches, n_in_channels, in_h, in_w = x.shape
        n_out_channels = weight.shape[0]

        out_h = (in_h + 2 * pad_h - (dil_h * (weight_h - 1) + 1)) // stride_h + 1
        out_w = (in_w + 2 * pad_w - (dil_w * (weight_w - 1) + 1)) // stride_w + 1

        n_offset_grps = offset.shape[1] // (2 * weight_h * weight_w)
        in_c_per_offset_grp = n_in_channels // n_offset_grps

        n_weight_grps = n_in_channels // weight.shape[1]
        in_c_per_weight_grp = weight.shape[1]
        out_c_per_weight_grp = n_out_channels // n_weight_grps

        out = torch.zeros(n_batches, n_out_channels, out_h, out_w, device=x.device, dtype=x.dtype)
        for b in range(n_batches):
            for c_out in range(n_out_channels):
                for i in range(out_h):
                    for j in range(out_w):
                        for di in range(weight_h):
                            for dj in range(weight_w):
                                for c in range(in_c_per_weight_grp):
                                    weight_grp = c_out // out_c_per_weight_grp
                                    c_in = weight_grp * in_c_per_weight_grp + c

                                    offset_grp = c_in // in_c_per_offset_grp
                                    mask_idx = offset_grp * (weight_h * weight_w) + di * weight_w + dj
                                    offset_idx = 2 * mask_idx

                                    pi = stride_h * i - pad_h + dil_h * di + offset[b, offset_idx, i, j]
                                    pj = stride_w * j - pad_w + dil_w * dj + offset[b, offset_idx + 1, i, j]

                                    mask_value = 1.0
                                    if mask is not None:
                                        mask_value = mask[b, mask_idx, i, j]

                                    out[b, c_out, i, j] += (mask_value * weight[c_out, c, di, dj] *
                                                            bilinear_interpolate(x[b, c_in, :, :], pi, pj))
        out += bias.view(1, n_out_channels, 1, 1)
        return out

    @lru_cache(maxsize=None)
    def get_fn_args(self, device, contiguous, batch_sz, dtype):
        n_in_channels = 6
        n_out_channels = 2
        n_weight_grps = 2
        n_offset_grps = 3

        stride = (2, 1)
        pad = (1, 0)
        dilation = (2, 1)

        stride_h, stride_w = stride
        pad_h, pad_w = pad
        dil_h, dil_w = dilation
        weight_h, weight_w = (3, 2)
        in_h, in_w = (5, 4)

        out_h = (in_h + 2 * pad_h - (dil_h * (weight_h - 1) + 1)) // stride_h + 1
        out_w = (in_w + 2 * pad_w - (dil_w * (weight_w - 1) + 1)) // stride_w + 1

        x = torch.rand(batch_sz, n_in_channels, in_h, in_w, device=device, dtype=dtype, requires_grad=True)

        offset = torch.randn(batch_sz, n_offset_grps * 2 * weight_h * weight_w, out_h, out_w,
                             device=device, dtype=dtype, requires_grad=True)

        mask = torch.randn(batch_sz, n_offset_grps * weight_h * weight_w, out_h, out_w,
                           device=device, dtype=dtype, requires_grad=True)

        weight = torch.randn(n_out_channels, n_in_channels // n_weight_grps, weight_h, weight_w,
                             device=device, dtype=dtype, requires_grad=True)

        bias = torch.randn(n_out_channels, device=device, dtype=dtype, requires_grad=True)

        if not contiguous:
            x = x.permute(0, 1, 3, 2).contiguous().permute(0, 1, 3, 2)
            offset = offset.permute(1, 3, 0, 2).contiguous().permute(2, 0, 3, 1)
            mask = mask.permute(1, 3, 0, 2).contiguous().permute(2, 0, 3, 1)
            weight = weight.permute(3, 2, 0, 1).contiguous().permute(2, 3, 1, 0)

        return x, weight, offset, mask, bias, stride, pad, dilation

    @pytest.mark.parametrize('device', cpu_and_gpu())
    @pytest.mark.parametrize('contiguous', (True, False))
    @pytest.mark.parametrize('batch_sz', (0, 33))
    def test_forward(self, device, contiguous, batch_sz, dtype=None):
        dtype = dtype or self.dtype
        x, _, offset, mask, _, stride, padding, dilation = self.get_fn_args(device, contiguous, batch_sz, dtype)
        in_channels = 6
        out_channels = 2
        kernel_size = (3, 2)
        groups = 2
        tol = 2e-3 if dtype is torch.half else 1e-5

        layer = ops.DeformConv2d(in_channels, out_channels, kernel_size, stride=stride, padding=padding,
                                 dilation=dilation, groups=groups).to(device=x.device, dtype=dtype)
        res = layer(x, offset, mask)

        weight = layer.weight.data
        bias = layer.bias.data
        expected = self.expected_fn(x, weight, offset, mask, bias, stride=stride, padding=padding, dilation=dilation)

        torch.testing.assert_close(
            res.to(expected), expected, rtol=tol, atol=tol, msg='\nres:\n{}\nexpected:\n{}'.format(res, expected)
        )

        # no modulation test
        res = layer(x, offset)
        expected = self.expected_fn(x, weight, offset, None, bias, stride=stride, padding=padding, dilation=dilation)

        torch.testing.assert_close(
            res.to(expected), expected, rtol=tol, atol=tol, msg='\nres:\n{}\nexpected:\n{}'.format(res, expected)
        )

    def test_wrong_sizes(self):
        in_channels = 6
        out_channels = 2
        kernel_size = (3, 2)
        groups = 2
        x, _, offset, mask, _, stride, padding, dilation = self.get_fn_args('cpu', contiguous=True,
                                                                            batch_sz=10, dtype=self.dtype)
        layer = ops.DeformConv2d(in_channels, out_channels, kernel_size, stride=stride, padding=padding,
                                 dilation=dilation, groups=groups)
        with pytest.raises(RuntimeError, match="the shape of the offset"):
            wrong_offset = torch.rand_like(offset[:, :2])
            layer(x, wrong_offset)

        with pytest.raises(RuntimeError, match=r'mask.shape\[1\] is not valid'):
            wrong_mask = torch.rand_like(mask[:, :2])
            layer(x, offset, wrong_mask)

    @pytest.mark.parametrize('device', cpu_and_gpu())
    @pytest.mark.parametrize('contiguous', (True, False))
    @pytest.mark.parametrize('batch_sz', (0, 33))
    def test_backward(self, device, contiguous, batch_sz):
        x, weight, offset, mask, bias, stride, padding, dilation = self.get_fn_args(device, contiguous,
                                                                                    batch_sz, self.dtype)

        def func(x_, offset_, mask_, weight_, bias_):
            return ops.deform_conv2d(x_, offset_, weight_, bias_, stride=stride,
                                     padding=padding, dilation=dilation, mask=mask_)

        gradcheck(func, (x, offset, mask, weight, bias), nondet_tol=1e-5, fast_mode=True)

        def func_no_mask(x_, offset_, weight_, bias_):
            return ops.deform_conv2d(x_, offset_, weight_, bias_, stride=stride,
                                     padding=padding, dilation=dilation, mask=None)

        gradcheck(func_no_mask, (x, offset, weight, bias), nondet_tol=1e-5, fast_mode=True)

        @torch.jit.script
        def script_func(x_, offset_, mask_, weight_, bias_, stride_, pad_, dilation_):
            # type:(Tensor, Tensor, Tensor, Tensor, Tensor, Tuple[int, int], Tuple[int, int], Tuple[int, int])->Tensor
            return ops.deform_conv2d(x_, offset_, weight_, bias_, stride=stride_,
                                     padding=pad_, dilation=dilation_, mask=mask_)

        gradcheck(lambda z, off, msk, wei, bi: script_func(z, off, msk, wei, bi, stride, padding, dilation),
                  (x, offset, mask, weight, bias), nondet_tol=1e-5, fast_mode=True)

        @torch.jit.script
        def script_func_no_mask(x_, offset_, weight_, bias_, stride_, pad_, dilation_):
            # type:(Tensor, Tensor, Tensor, Tensor, Tuple[int, int], Tuple[int, int], Tuple[int, int])->Tensor
            return ops.deform_conv2d(x_, offset_, weight_, bias_, stride=stride_,
                                     padding=pad_, dilation=dilation_, mask=None)

        gradcheck(lambda z, off, wei, bi: script_func_no_mask(z, off, wei, bi, stride, padding, dilation),
                  (x, offset, weight, bias), nondet_tol=1e-5, fast_mode=True)

    @needs_cuda
    @pytest.mark.parametrize('contiguous', (True, False))
    def test_compare_cpu_cuda_grads(self, contiguous):
        # Test from https://github.com/pytorch/vision/issues/2598
        # Run on CUDA only

        # compare grads computed on CUDA with grads computed on CPU
        true_cpu_grads = None

        init_weight = torch.randn(9, 9, 3, 3, requires_grad=True)
        img = torch.randn(8, 9, 1000, 110)
        offset = torch.rand(8, 2 * 3 * 3, 1000, 110)
        mask = torch.rand(8, 3 * 3, 1000, 110)

        if not contiguous:
            img = img.permute(0, 1, 3, 2).contiguous().permute(0, 1, 3, 2)
            offset = offset.permute(1, 3, 0, 2).contiguous().permute(2, 0, 3, 1)
            mask = mask.permute(1, 3, 0, 2).contiguous().permute(2, 0, 3, 1)
            weight = init_weight.permute(3, 2, 0, 1).contiguous().permute(2, 3, 1, 0)
        else:
            weight = init_weight

        for d in ["cpu", "cuda"]:

            out = ops.deform_conv2d(img.to(d), offset.to(d), weight.to(d), padding=1, mask=mask.to(d))
            out.mean().backward()
            if true_cpu_grads is None:
                true_cpu_grads = init_weight.grad
                assert true_cpu_grads is not None
            else:
                assert init_weight.grad is not None
                res_grads = init_weight.grad.to("cpu")
                torch.testing.assert_close(true_cpu_grads, res_grads)

    @needs_cuda
    @pytest.mark.parametrize('batch_sz', (0, 33))
    @pytest.mark.parametrize('dtype', (torch.float, torch.half))
    def test_autocast(self, batch_sz, dtype):
        with torch.cuda.amp.autocast():
            self.test_forward(torch.device("cuda"), contiguous=False, batch_sz=batch_sz, dtype=dtype)


class TestFrozenBNT:
    def test_frozenbatchnorm2d_repr(self):
        num_features = 32
        eps = 1e-5
        t = ops.misc.FrozenBatchNorm2d(num_features, eps=eps)

        # Check integrity of object __repr__ attribute
        expected_string = f"FrozenBatchNorm2d({num_features}, eps={eps})"
        assert repr(t) == expected_string

    def test_frozenbatchnorm2d_eps(self):
        sample_size = (4, 32, 28, 28)
        x = torch.rand(sample_size)
        state_dict = dict(weight=torch.rand(sample_size[1]),
                          bias=torch.rand(sample_size[1]),
                          running_mean=torch.rand(sample_size[1]),
                          running_var=torch.rand(sample_size[1]),
                          num_batches_tracked=torch.tensor(100))

        # Check that default eps is equal to the one of BN
        fbn = ops.misc.FrozenBatchNorm2d(sample_size[1])
        fbn.load_state_dict(state_dict, strict=False)
        bn = torch.nn.BatchNorm2d(sample_size[1]).eval()
        bn.load_state_dict(state_dict)
        # Difference is expected to fall in an acceptable range
        torch.testing.assert_close(fbn(x), bn(x), rtol=1e-5, atol=1e-6)

        # Check computation for eps > 0
        fbn = ops.misc.FrozenBatchNorm2d(sample_size[1], eps=1e-5)
        fbn.load_state_dict(state_dict, strict=False)
        bn = torch.nn.BatchNorm2d(sample_size[1], eps=1e-5).eval()
        bn.load_state_dict(state_dict)
        torch.testing.assert_close(fbn(x), bn(x), rtol=1e-5, atol=1e-6)

    def test_frozenbatchnorm2d_n_arg(self):
        """Ensure a warning is thrown when passing `n` kwarg
        (remove this when support of `n` is dropped)"""
        with pytest.warns(DeprecationWarning):
            ops.misc.FrozenBatchNorm2d(32, eps=1e-5, n=32)


class TestBoxConversion:
    def _get_box_sequences():
        # Define here the argument type of `boxes` supported by region pooling operations
        box_tensor = torch.tensor([[0, 0, 0, 100, 100], [1, 0, 0, 100, 100]], dtype=torch.float)
        box_list = [torch.tensor([[0, 0, 100, 100]], dtype=torch.float),
                    torch.tensor([[0, 0, 100, 100]], dtype=torch.float)]
        box_tuple = tuple(box_list)
        return box_tensor, box_list, box_tuple

    @pytest.mark.parametrize('box_sequence', _get_box_sequences())
    def test_check_roi_boxes_shape(self, box_sequence):
        # Ensure common sequences of tensors are supported
        ops._utils.check_roi_boxes_shape(box_sequence)

    @pytest.mark.parametrize('box_sequence', _get_box_sequences())
    def test_convert_boxes_to_roi_format(self, box_sequence):
        # Ensure common sequences of tensors yield the same result
        ref_tensor = None
        if ref_tensor is None:
            ref_tensor = box_sequence
        else:
            assert_equal(ref_tensor, ops._utils.convert_boxes_to_roi_format(box_sequence))


class TestBox:
    def test_bbox_same(self):
        box_tensor = torch.tensor([[0, 0, 100, 100], [0, 0, 0, 0],
                                  [10, 15, 30, 35], [23, 35, 93, 95]], dtype=torch.float)

        exp_xyxy = torch.tensor([[0, 0, 100, 100], [0, 0, 0, 0],
                                [10, 15, 30, 35], [23, 35, 93, 95]], dtype=torch.float)

        assert exp_xyxy.size() == torch.Size([4, 4])
        assert_equal(ops.box_convert(box_tensor, in_fmt="xyxy", out_fmt="xyxy"), exp_xyxy)
        assert_equal(ops.box_convert(box_tensor, in_fmt="xywh", out_fmt="xywh"), exp_xyxy)
        assert_equal(ops.box_convert(box_tensor, in_fmt="cxcywh", out_fmt="cxcywh"), exp_xyxy)

    def test_bbox_xyxy_xywh(self):
        # Simple test convert boxes to xywh and back. Make sure they are same.
        # box_tensor is in x1 y1 x2 y2 format.
        box_tensor = torch.tensor([[0, 0, 100, 100], [0, 0, 0, 0],
                                  [10, 15, 30, 35], [23, 35, 93, 95]], dtype=torch.float)
        exp_xywh = torch.tensor([[0, 0, 100, 100], [0, 0, 0, 0],
                                [10, 15, 20, 20], [23, 35, 70, 60]], dtype=torch.float)

        assert exp_xywh.size() == torch.Size([4, 4])
        box_xywh = ops.box_convert(box_tensor, in_fmt="xyxy", out_fmt="xywh")
        assert_equal(box_xywh, exp_xywh)

        # Reverse conversion
        box_xyxy = ops.box_convert(box_xywh, in_fmt="xywh", out_fmt="xyxy")
        assert_equal(box_xyxy, box_tensor)

    def test_bbox_xyxy_cxcywh(self):
        # Simple test convert boxes to xywh and back. Make sure they are same.
        # box_tensor is in x1 y1 x2 y2 format.
        box_tensor = torch.tensor([[0, 0, 100, 100], [0, 0, 0, 0],
                                  [10, 15, 30, 35], [23, 35, 93, 95]], dtype=torch.float)
        exp_cxcywh = torch.tensor([[50, 50, 100, 100], [0, 0, 0, 0],
                                  [20, 25, 20, 20], [58, 65, 70, 60]], dtype=torch.float)

        assert exp_cxcywh.size() == torch.Size([4, 4])
        box_cxcywh = ops.box_convert(box_tensor, in_fmt="xyxy", out_fmt="cxcywh")
        assert_equal(box_cxcywh, exp_cxcywh)

        # Reverse conversion
        box_xyxy = ops.box_convert(box_cxcywh, in_fmt="cxcywh", out_fmt="xyxy")
        assert_equal(box_xyxy, box_tensor)

    def test_bbox_xywh_cxcywh(self):
        box_tensor = torch.tensor([[0, 0, 100, 100], [0, 0, 0, 0],
                                  [10, 15, 20, 20], [23, 35, 70, 60]], dtype=torch.float)

        # This is wrong
        exp_cxcywh = torch.tensor([[50, 50, 100, 100], [0, 0, 0, 0],
                                  [20, 25, 20, 20], [58, 65, 70, 60]], dtype=torch.float)

        assert exp_cxcywh.size() == torch.Size([4, 4])
        box_cxcywh = ops.box_convert(box_tensor, in_fmt="xywh", out_fmt="cxcywh")
        assert_equal(box_cxcywh, exp_cxcywh)

        # Reverse conversion
        box_xywh = ops.box_convert(box_cxcywh, in_fmt="cxcywh", out_fmt="xywh")
        assert_equal(box_xywh, box_tensor)

    @pytest.mark.parametrize('inv_infmt', ["xwyh", "cxwyh"])
    @pytest.mark.parametrize('inv_outfmt', ["xwcx", "xhwcy"])
    def test_bbox_invalid(self, inv_infmt, inv_outfmt):
        box_tensor = torch.tensor([[0, 0, 100, 100], [0, 0, 0, 0],
                                  [10, 15, 20, 20], [23, 35, 70, 60]], dtype=torch.float)

        with pytest.raises(ValueError):
            ops.box_convert(box_tensor, inv_infmt, inv_outfmt)

    def test_bbox_convert_jit(self):
        box_tensor = torch.tensor([[0, 0, 100, 100], [0, 0, 0, 0],
                                  [10, 15, 30, 35], [23, 35, 93, 95]], dtype=torch.float)

        scripted_fn = torch.jit.script(ops.box_convert)
        TOLERANCE = 1e-3

        box_xywh = ops.box_convert(box_tensor, in_fmt="xyxy", out_fmt="xywh")
        scripted_xywh = scripted_fn(box_tensor, 'xyxy', 'xywh')
        torch.testing.assert_close(scripted_xywh, box_xywh, rtol=0.0, atol=TOLERANCE)

        box_cxcywh = ops.box_convert(box_tensor, in_fmt="xyxy", out_fmt="cxcywh")
        scripted_cxcywh = scripted_fn(box_tensor, 'xyxy', 'cxcywh')
        torch.testing.assert_close(scripted_cxcywh, box_cxcywh, rtol=0.0, atol=TOLERANCE)


class TestBoxArea:
    def test_box_area(self):
        def area_check(box, expected, tolerance=1e-4):
            out = ops.box_area(box)
            torch.testing.assert_close(out, expected, rtol=0.0, check_dtype=False, atol=tolerance)

        # Check for int boxes
        for dtype in [torch.int8, torch.int16, torch.int32, torch.int64]:
            box_tensor = torch.tensor([[0, 0, 100, 100], [0, 0, 0, 0]], dtype=dtype)
            expected = torch.tensor([10000, 0])
            area_check(box_tensor, expected)

        # Check for float32 and float64 boxes
        for dtype in [torch.float32, torch.float64]:
            box_tensor = torch.tensor([[285.3538, 185.5758, 1193.5110, 851.4551],
                                       [285.1472, 188.7374, 1192.4984, 851.0669],
                                       [279.2440, 197.9812, 1189.4746, 849.2019]], dtype=dtype)
            expected = torch.tensor([604723.0806, 600965.4666, 592761.0085], dtype=torch.float64)
            area_check(box_tensor, expected, tolerance=0.05)

        # Check for float16 box
        box_tensor = torch.tensor([[285.25, 185.625, 1194.0, 851.5],
                                   [285.25, 188.75, 1192.0, 851.0],
                                   [279.25, 198.0, 1189.0, 849.0]], dtype=torch.float16)
        expected = torch.tensor([605113.875, 600495.1875, 592247.25])
        area_check(box_tensor, expected)


class TestBoxIou:
    def test_iou(self):
        def iou_check(box, expected, tolerance=1e-4):
            out = ops.box_iou(box, box)
            torch.testing.assert_close(out, expected, rtol=0.0, check_dtype=False, atol=tolerance)

        # Check for int boxes
        for dtype in [torch.int16, torch.int32, torch.int64]:
            box = torch.tensor([[0, 0, 100, 100], [0, 0, 50, 50], [200, 200, 300, 300]], dtype=dtype)
            expected = torch.tensor([[1.0, 0.25, 0.0], [0.25, 1.0, 0.0], [0.0, 0.0, 1.0]])
            iou_check(box, expected)

        # Check for float boxes
        for dtype in [torch.float16, torch.float32, torch.float64]:
            box_tensor = torch.tensor([[285.3538, 185.5758, 1193.5110, 851.4551],
                                       [285.1472, 188.7374, 1192.4984, 851.0669],
                                       [279.2440, 197.9812, 1189.4746, 849.2019]], dtype=dtype)
            expected = torch.tensor([[1.0, 0.9933, 0.9673], [0.9933, 1.0, 0.9737], [0.9673, 0.9737, 1.0]])
            iou_check(box_tensor, expected, tolerance=0.002 if dtype == torch.float16 else 1e-4)


class TestGenBoxIou:
    def test_gen_iou(self):
        def gen_iou_check(box, expected, tolerance=1e-4):
            out = ops.generalized_box_iou(box, box)
            torch.testing.assert_close(out, expected, rtol=0.0, check_dtype=False, atol=tolerance)

        # Check for int boxes
        for dtype in [torch.int16, torch.int32, torch.int64]:
            box = torch.tensor([[0, 0, 100, 100], [0, 0, 50, 50], [200, 200, 300, 300]], dtype=dtype)
            expected = torch.tensor([[1.0, 0.25, -0.7778], [0.25, 1.0, -0.8611], [-0.7778, -0.8611, 1.0]])
            gen_iou_check(box, expected)

        # Check for float boxes
        for dtype in [torch.float16, torch.float32, torch.float64]:
            box_tensor = torch.tensor([[285.3538, 185.5758, 1193.5110, 851.4551],
                                       [285.1472, 188.7374, 1192.4984, 851.0669],
                                       [279.2440, 197.9812, 1189.4746, 849.2019]], dtype=dtype)
            expected = torch.tensor([[1.0, 0.9933, 0.9673], [0.9933, 1.0, 0.9737], [0.9673, 0.9737, 1.0]])
            gen_iou_check(box_tensor, expected, tolerance=0.002 if dtype == torch.float16 else 1e-3)


if __name__ == '__main__':
    pytest.main([__file__])