hiSSKJr SSKrSSKrSSKrSSKrSSKrSSKrSSK r SSK J s J r SSKJr "SS\R 5rS"S#SjjrS$S%SjjrS&SjrS rS rS'S jrS rS rS(S)SjjrS*S+SjjrSrSrSrSrSr Sr!Sr"S,S-Sjjr#Sr$S.S/Sjjr%Sr&S0S1Sjjr'S2S3Sjjr(Sr)S4S5Sjjr*S6Sjr+S r,S!r-g)7) annotationsN) NOT_MACOS14c>\rSrSrSrS S SjjrSrSrSrSr S r g) Profilea Ultralytics Profile class for timing code execution. Use as a decorator with @Profile() or as a context manager with 'with Profile():'. Provides accurate timing measurements with CUDA synchronization support for GPU operations. Attributes: t (float): Accumulated time in seconds. device (torch.device): Device used for model inference. cuda (bool): Whether CUDA is being used for timing synchronization. Examples: Use as a context manager to time code execution >>> with Profile(device=device) as dt: ... pass # slow operation here >>> print(dt) # prints "Elapsed time is 9.5367431640625e-07 s" Use as a decorator to time function execution >>> @Profile() ... def slow_function(): ... time.sleep(0.1) Nc~XlX l[U=(a [U5R S55Ulg)z Initialize the Profile class. Args: t (float): Initial accumulated time in seconds. device (torch.device, optional): Device used for model inference to enable CUDA synchronization. cudaN)tdeviceboolstr startswithr )selfr r s O/home/james-whalen/.local/lib/python3.13/site-packages/ultralytics/utils/ops.py__init__Profile.__init__*s. BCK$:$:6$BC c0UR5UlU$)z Start timing.)timestartrs r __enter__Profile.__enter__6sYY[  rcUR5UR- UlU=RUR- slg)z Stop timing.N)rrdtr )rtypevalue tracebacks r__exit__Profile.__exit__;s*))+ * $''rc"SURS3$)zIReturn a human-readable string representing the accumulated elapsed time.zElapsed time is z s)r rs r__str__Profile.__str__@s!$&&,,rcUR(a)[RRUR5 [R "5$)z9Get current time with CUDA synchronization if applicable.)r torch synchronizer r perf_counterrs rr Profile.timeDs. 99 JJ " "4;; /  ""r)r r rrr )N)r floatr ztorch.device | None) __name__ __module__ __qualname____firstlineno____doc__rrrr"r__static_attributes__rrrrs!. D  -#rrcURup4[R"UR5S:UR5S:UR 5U:UR 5U:/5R 5S:a$UR SU5nUR SU5nUS:US:-X1:*-XB:*-nX5nXEn[U5(a[[R"UR5UR5UR 5UR 5/URS9$[R"SURS9$)a Convert segment coordinates to bounding box coordinates. Converts a single segment label to a box label by finding the minimum and maximum x and y coordinates. Applies inside-image constraint and clips coordinates when necessary. Args: segment (torch.Tensor): Segment coordinates in format (N, 2) where N is number of points. width (int): Width of the image in pixels. height (int): Height of the image in pixels. Returns: (np.ndarray): Bounding box coordinates in xyxy format [x1, y1, x2, y2]. rdtype) Tnparrayminmaxsumclipanyr5zeros)segmentwidthheightxyinsides r segment2boxrFKs 99DA xx1aeegk1557U?AEEGfNN5R]R]^`RaehRhklRlsu[R[R[R [R 4O=[R[R[R[R 4upp4TSSS24nU4Sj[SS55upgnU"U5U"U5pUS- U -US- U -/n U*S- U -US- U -/n U"U S5n U"U S5n X[-U -n X[-U - nX[- U - nX[- U -nU"XUU/S5$)a~ Convert batched Oriented Bounding Boxes (OBB) from [xywh, rotation] to [xy1, xy2, xy3, xy4] format. Args: x (np.ndarray | torch.Tensor): Boxes in [cx, cy, w, h, rotation] format with shape (N, 5) or (B, N, 5). Rotation values should be in radians from 0 to pi/2. Returns: (np.ndarray | torch.Tensor): Converted corner points with shape (N, 4, 2) or (B, N, 4, 2). .NrIc3:># UHnTSXS-24v M g7f).rHNr1).0irCs r !xywhr2xyxyxyxy..s :k1S!!e)^$ksrn) r\r%r]cossincatstackr8rvrange)rCrrrrctrrirhr cos_value sin_valuevec1vec2pt1pt2pt3pt4s` rxywhr2xyxyxyxyrs# a & & EIIuyy%++6ffbffbnnbhh 7Cc C!G*C:eAqk:KA%u:s5zy EI q1uy0 1D BFY A 1 2D tR=D tR=D *t C *t C *t C *t C #C%r **rc[U[R5(aUR5O[R "U5nUSUS-US'USUS-US'U$)a Convert bounding box from [x1, y1, w, h] to [x1, y1, x2, y2] where xy1=top-left, xy2=bottom-right. Args: x (np.ndarray | torch.Tensor): Input bounding box coordinates. Returns: (np.ndarray | torch.Tensor): Bounding box coordinates in xyxy format. rMrKrNrLrrs r ltwh2xyxyrrrc /nUH^nURup4URUR5UR5UR5UR5/5 M` [ [ R "U55$)a  Convert segment labels to box labels, i.e. (cls, xy1, xy2, ...) to (cls, xywh). Args: segments (list): List of segments where each segment is a list of points, each point is [x, y] coordinates. Returns: (np.ndarray): Bounding box coordinates in xywh format. )r7rr:r;rr8r9)segmentsrRsrCrDs rsegments2boxesrs` E ss aeegquuw9: RXXe_ %%rc [U5GHIup#[U5U:XaM[R"X3SS2SS244SS9n[R"S[U5S- [U5U:aU[U5- OU5n[R "[U55n[U5U:a,[R "U[R"XE5U5OUn[R"[S5Vs/sH!n[R"XEUSS2U45PM# sn[RS9RSS5RUW'GML U$s snf)a Resample segments to n points each using linear interpolation. Args: segments (list): List of (N, 2) arrays where N is the number of points in each segment. n (int): Number of points to resample each segment to. Returns: (list): Resampled segments with n points each. rrHNrorIr4rn) enumeraterqr8rvlinspacearangeinsert searchsortedrinterpfloat32rr7)rnrrrCxps rresample_segmentsrs(# q6Q;  NNA1ay> 2 KK3q6A:SVaZq3q6zQ G YYs1v 8;A BIIa/ 4 NNuQxHx!BIIaQq!tW5xHPRPZPZ [ c cdegi j l l  $ OIs<(E cURup#nUS:a\[UR5R55H/unupgpSXSU24'SXU S24'SXSS2SU24'SXSS2US24'M1 U$[R "USS2SS2S4SS5upgp[R "X@RURS9SSSS24n [R "X0RURS9SSS2S4n X U:X:-X:-X:--$)z Crop masks to bounding box regions. Args: masks (torch.Tensor): Masks with shape (N, H, W). boxes (torch.Tensor): Bounding box coordinates with shape (N, 4) in relative point form. Returns: (torch.Tensor): Cropped masks. 2rNr6rHr) rgrrOr^r%chunkrr r5) rwrRrrhrirrrrrrcs r crop_maskrskkGA!2v#,U[[]->->-@#A AESbS&MERS&M EQ)  EQ)  $B  U1a:%61= LL<    rcURupEnXR5RUS5-RSXV5n[USU5Sn[ Xr5nUR S5R 5$)a Apply masks to bounding boxes using mask head output with native upsampling. Args: protos (torch.Tensor): Mask prototypes with shape (mask_dim, mask_h, mask_w). masks_in (torch.Tensor): Mask coefficients with shape (N, mask_dim) where N is number of masks after NMS. bboxes (torch.Tensor): Bounding boxes with shape (N, 4) where N is number of masks after NMS. shape (tuple): Input image size as (height, width). Returns: (torch.Tensor): Binary mask tensor with shape (H, W, N). rnNrr))rgr*r scale_masksrrr)rrrrgrrrrws rprocess_mask_nativer3so IA2  ++Ar2 2 8 8R DE d U +A .E e $E 99S>    rcURSSup4[X1S- XAS- 5nXASU-- nX1SU-- nU(a US-nUS-nU(a.[[US- 55[[US- 554OSupU[[US-55- n U[[US-55- n [R "USX2X24USS 9$) a. Rescale segment masks to target shape. Args: masks (torch.Tensor): Masks with shape (N, C, H, W). shape (tuple): Target height and width as (height, width). padding (bool): Whether masks are based on YOLO-style augmented images with padding. Returns: (torch.Tensor): Rescaled masks. rINrrHrJ)rr.rr)rgr:r^rOrr) rwrgrUrrrWrrrrrrs rrrGs[[_FB r!H}b8m ,D qD E qD E    FMU53;'(#eECK.@*ABSYIC #eECK() )F U53;'( (E ==sCJ :;U TTrcfUSSupgUc,USSup[X- X- 5n XU -- S- XU -- S- 4n O USSn USn U(a US==U S-ss'US==U S-ss'US==U -ss'US==U -ss'[X5nU(aUS==U-ss'US==U-ss'U$)aO Rescale segment coordinates from img1_shape to img0_shape. Args: img1_shape (tuple): Source image shape as HWC or HW (supports both). coords (torch.Tensor): Coordinates to scale with shape (N, 2). img0_shape (tuple): Image 0 shape as HWC or HW (supports both). ratio_pad (tuple, optional): Ratio and padding values as ((ratio_h, ratio_w), (pad_h, pad_w)). normalize (bool): Whether to normalize coordinates to range [0, 1]. padding (bool): Whether coordinates are based on YOLO-style augmented images with padding. Returns: (torch.Tensor): Scaled coordinates. NrIrrHrKrL)r:rl) rQrkrSrT normalizerUimg0_himg0_wimg1_himg1_wrWr~s r scale_coordsr `s ^NF#BQ6?FO4 %*Vtm-Cq,HH|Alv#a& v#a&  6NdN 6NdN  ,Fv& v&  Mrc.URSS9upp4nU[R-[RS- :n[R"XdU5n[R"XcU5nU[RS- -n[R "XXxU/SS9$)z Regularize rotated bounding boxes to range [0, pi/2]. Args: rboxes (torch.Tensor): Input rotated boxes with shape (N, 5) in xywhr format. Returns: (torch.Tensor): Regularized rotated boxes. rn)dimrI)unbindr_rr%wherer) rrCrDrirhr swapw_h_s rregularize_rboxesrs~MMbM)MA! tww;$''A+ %D Ta B Ta B TWWq[A ;;ba(b 11rc "SSKJn /nUR5R5R 5GHKn[ R "U[ R[ R5SnU(aUS:Xaa[U5S:a<[R"U"UVs/sHoDRSS5PM sn55OUSRSS5nOUS:Xae[R"U[R"UVs/sHn[U5PM sn5R55RSS5nO[R"S5nUR!UR#S 55 GMN U$s snfs snf) a  Convert masks to segments using contour detection. Args: masks (torch.Tensor): Binary masks with shape (batch_size, 160, 160). strategy (str): Segmentation strategy, either 'all' or 'largest'. Returns: (list): List of segment masks as float32 arrays. r)merge_multi_segmentallrHrnrIlargest)rrIr)ultralytics.data.converterrrrrrt findContours RETR_EXTERNALCHAIN_APPROX_SIMPLErqr8rvrr9argmaxr?rastype)rwstrategyrrrCrs rmasks2segmentsrs8?H ZZ\    % % '   Q 1 133J3J KA N 5 1vzNN#6RS7TRSQ "a8HRS7T#UV1b!, Y&HHQrxx(;AQ(;<CCEFGOOPRTUV A+,( O8U )<+]r})patternreplstring)resub)rs r clean_strr+s 66>SQR SSrc[U[R5(a#[R"U[RS9$[ R"U[ RS9$)zSCreate empty torch.Tensor or np.ndarray with same shape as input and float32 dtype.r4)r\r%r]rrr8)rCs rrrsJ5?q%,,4O4O%--0UWUbUbcdlnlvlvUwr)r-)rAr^rBr^)NTF)rUr rVr )rCr^)N)r-r-rr)rir^rhr^rr^rr^)r-r-Fr))rir^rhr^r=r rr*)i)rr^)F)rr )T)rUr )NFT)rr rUr )r)rr )r#z torch.Tensorreturnz np.ndarray). __future__r contextlibr_r)rrtrr8r%torch.nn.functionalnn functionalrultralytics.utilsrContextDecoratorrrFrZrbrPrlrrrrrrrrrrrrrrrrrr rrr$r+rr1rrr6s#  )6#j))6#r  &"0E4!8!(U2!H2&> ] Tr