import torch

def rgb2hsv_torch(rgb: torch.Tensor) -> torch.Tensor:
    cmax, cmax_idx = torch.max(rgb, dim=1, keepdim=True)
    cmin = torch.min(rgb, dim=1, keepdim=True)[0]
    delta = cmax - cmin
    hsv_h = torch.empty_like(rgb[:, 0:1, :, :])
    cmax_idx[delta == 0] = 3
    hsv_h[cmax_idx == 0] = (((rgb[:, 1:2] - rgb[:, 2:3]) / delta) % 6)[cmax_idx == 0]
    hsv_h[cmax_idx == 1] = (((rgb[:, 2:3] - rgb[:, 0:1]) / delta) + 2)[cmax_idx == 1]
    hsv_h[cmax_idx == 2] = (((rgb[:, 0:1] - rgb[:, 1:2]) / delta) + 4)[cmax_idx == 2]
    hsv_h[cmax_idx == 3] = 0.
    hsv_h /= 6.
    hsv_s = torch.where(cmax == 0, torch.tensor(0.).type_as(rgb), delta / cmax)
    hsv_v = cmax
    return torch.cat([hsv_h, hsv_s, hsv_v], dim=1)


def hsv2rgb_torch(hsv: torch.Tensor) -> torch.Tensor:
    hsv_h, hsv_s, hsv_l = hsv[:, 0:1], hsv[:, 1:2], hsv[:, 2:3]
    _c = hsv_l * hsv_s
    _x = _c * (- torch.abs(hsv_h * 6. % 2. - 1) + 1.)
    _m = hsv_l - _c
    _o = torch.zeros_like(_c)
    idx = (hsv_h * 6.).type(torch.uint8)
    idx = (idx % 6).expand(-1, 3, -1, -1)
    rgb = torch.empty_like(hsv)
    rgb[idx == 0] = torch.cat([_c, _x, _o], dim=1)[idx == 0]
    rgb[idx == 1] = torch.cat([_x, _c, _o], dim=1)[idx == 1]
    rgb[idx == 2] = torch.cat([_o, _c, _x], dim=1)[idx == 2]
    rgb[idx == 3] = torch.cat([_o, _x, _c], dim=1)[idx == 3]
    rgb[idx == 4] = torch.cat([_x, _o, _c], dim=1)[idx == 4]
    rgb[idx == 5] = torch.cat([_c, _o, _x], dim=1)[idx == 5]
    rgb += _m
    return rgb