a d| @s"ddlmZmZmZmZmZddlZdeeedddZ dde dfde d e feej e ee e fee fee ee e feeej ejej d d d Zdej eeee e fdd ddZdej eeee e feeddddZdeej e efeeej eejej dddZdS))ListOptionalTupleUnioncastNZ batch_sizeZ same_on_batchcCsLt|tur|dks$td|d|durHt|turHtd|ddS)z*Valid batch_size and same_on_batch params.rz.`batch_size` shall be a positive integer. Got .Nz&`same_on_batch` shall be boolean. Got )typeintAssertionErrorboolrr s/var/www/html/stable-diffusion-webui/venv/lib/python3.9/site-packages/kornia/augmentation/utils/param_validation.py_common_param_checksrginfjointcpu)factornamecenterboundscheckdevicedtypereturncCst|tjstj|||d}|dkr|dkrDtd|d||dtjddg|j|jd|}| |d|dj ||d}ntj |||d}|d ur|d krt |||n&|d krt |||ntd |d |S)z9Check inputs and compute the corresponding factor bounds.rrrzIf z9 is a single number number, it must be non negative. Got ?NrZsingularz methods 'z' not implemented.) isinstancetorchTensortensordim ValueErrorrepeatrrclamptoZ as_tensor_joint_range_check_singular_range_checkNotImplementedError)rrrrrrrZ factor_boundr r r _range_bounds  & r,) ranged_factorrrrcCs|durtdtdf}|dkrt|dkr|d|dkrP|d|dksjt|d|d|d |d|dkr|dkr|dksnt|d |d |nt|d |d |d dS)zGCheck if bounds[0] <= ranged_factor[0] <= ranged_factor[1] <= bounds[1]N-infrrrr out of bounds. Expected inside , got rz[0] should be smaller than z[1] got z7 should be a tensor with length 2 whose values between z. Got )floatr$lenr% TypeError)r-rrr r rr)2s 0r)F2d)r-rr skip_nonemodercCs|dkrd}n|dkrd}ntd||r:|dur:dS|durRtdtdf}|d krt||kr|D]<}|d |kr|d ksnnt|d |d |d qnnt|d|d|d|dS)zbCheck if bounds[0] <= ranged_factor[0] <= bounds[1] and bounds[0] <= ranged_factor[1] <= bounds[1]r4rZ3dz%'mode' shall be either 2d or 3d. Got Nr.rrrr/r0rz1 should be a float number or a tuple with length z whose values between z.Got )r%r1r$r2r3)r-rrr5r6Zdim_sizefr r rr*Bs$ r*) input_range target_sizerrrcCst|dg}t|szt|ttfrd|dkr>td|tj| |g||d|dd}qt|t t frt |dkrt|dttfrt|dttfrtj|||d|dd}nt|t t frt ||dkrt dd|Drtjdd |D||d}ndt|t t frZt ||dkrZt d d|DrZtj|||d}nt d |d |d |dn.ttj|}t |jdkst |jdkrt |dkr|dkrtd||dtjddg|j|jd}||dd}nt |jdkr0t |dkr0||dd}nxt |jdkr~t ||dkr~|dddtjddg|j|jd}n*|j|kr|}ntdt |d||S)aSGiven target_size, it will generate the corresponding (target_size, 2) range tensor for element-wise params. Example: >>> degree = torch.tensor([0.2, 0.3]) >>> _tuple_range_reader(degree, 3) # read degree for yaw, pitch and roll. tensor([[0.2000, 0.3000], [0.2000, 0.3000], [0.2000, 0.3000]]) rrzCIf input_range is only one number it must be a positive number. Gotrrcss|]}t|ttfVqdSN)r r1r .0xr r r z&_tuple_range_reader..cSsg|]}| |fqSr r )r=sr r r r@z'_tuple_range_reader..css|]}t|ttfVqdSr;)r tuplelistr<r r rr?r@z]If not pass a tensor, it must be float, (float, float) for isotropic operation or a tuple of z floats or z/ (float, float) for independent operation. Got rrrzDegrees must be a z; tensor for the degree range for independent operation.Got )r!SizeZ is_tensorr r1r r%r#r&rCrDr2allr3rr"shaperrZ unsqueeze)r9r:rrZ target_shapeZinput_range_tmpr r r_tuple_range_reader`s|      .  "   rI)N)N)NFr4)NN)typingrrrrrr!r r rr1rZget_default_dtyper"strrr,r)r*rCrIr r r rsL    $$ !