mindsponge.common

API Name	Description	Supported Platforms
`mindsponge.common.find_optimal_renaming`	Find optimal renaming for ground truth that maximizes LDDT.	`Ascend` `GPU`
`mindsponge.common.get_aligned_seq`	Align two protein fasta sequence.	`Ascend` `GPU`
`mindsponge.common.get_fasta_info`	Put in a pdb file and get fasta information from it.	`Ascend` `GPU`
`mindsponge.common.get_pdb_info`	get atom positions, residue index etc.	`Ascend` `GPU`
`mindsponge.common.make_atom14_positions`	The function of transforming sparse encoding method to densely encoding method.	`Ascend` `GPU`

mindsponge.common.geometry

API Name	Description	Supported Platforms
`mindsponge.common.apply_to_point`	Rotate and translate the input coordinates.	`Ascend` `GPU`
`mindsponge.common.initial_affine`	Initialize quaternion, rotation, translation of affine.	`Ascend` `GPU`
`mindsponge.common.invert_point`	The inverse transformation of a rigid body group transformation with respect to a point coordinate, that is, the inverse transformation of apply to point Make rotational translation changes on coordinates with the transpose of the rotation matrix $(x x, x y, x z, y x, y y, y z, z x, z y, z z)$ and the translation vector $(x, y, z)$ translation.	`Ascend` `GPU`
`mindsponge.common.invert_rigids`	Computes group inverse of rigid transformations.	`Ascend` `GPU`
`mindsponge.common.invert_rots`	Computes inverse of rotations $m$ .	`Ascend` `GPU`
`mindsponge.common.make_transform_from_reference`	Using GramSchmidt process to construct rotation and translation from given points.	`Ascend` `GPU`
`mindsponge.common.pre_compose`	Return a new QuatAffine which applies the transformation update first.	`Ascend` `GPU`
`mindsponge.common.quat_affine`	Create quat affine representations based on rots and trans.	`Ascend` `GPU`
`mindsponge.common.quat_multiply_by_vec`	Multiply a quaternion by a pure-vector quaternion.	`Ascend` `GPU`
`mindsponge.common.quat_to_rot`	Convert a normalized quaternion to a rotation matrix.	`Ascend` `GPU`
`mindsponge.common.quaternion_from_tensor`	Take the input 'tensor' $[(x x, x y, x z, y x, y y, y z, z z)]$ to get the new 'quaternion', 'rotation', 'translation'.	`Ascend` `GPU`
`mindsponge.common.quaternion_to_tensor`	Change quaternion to tensor.	`Ascend` `GPU`
`mindsponge.common.rigids_from_3_points`	Gram-Schmidt process.	`Ascend` `GPU`
`mindsponge.common.rigids_mul_rigids`	Change rigid $b$ from its local coordinate system to rigid $a$ local coordinate system, using rigid rotations and translations.	`Ascend` `GPU`
`mindsponge.common.rigids_mul_rots`	Numpy version of getting results rigid $x$ multiply rotations $\vec{y}$ .	`Ascend` `GPU`
`mindsponge.common.rigids_mul_vecs`	Transform vector $\vec{v}$ to rigid' local coordinate system.	`Ascend` `GPU`
`mindsponge.common.rots_expand_dims`	Adds an additional dimension to rots at the given axis.	`Ascend` `GPU`
`mindsponge.common.rots_from_tensor`	Amortize and split the 3*3 rotation matrix corresponding to the last two axes of input Tensor	`Ascend` `GPU`
`mindsponge.common.rots_from_two_vecs`	Put in two vectors $\vec{a} = (a_{x}, a_{y}, a_{z})$ and $\vec{b} = (b_{x}, b_{y}, b_{z})$ .	`Ascend` `GPU`
`mindsponge.common.rots_mul_rots`	Get result of rotation matrix x multiply rotation matrix y.	`Ascend` `GPU`
`mindsponge.common.rots_mul_vecs`	Apply rotations $\vec{m} = (m_{0}, m_{1}, m_{2}, m_{3}, m_{4}, m_{5}, m_{6}, m_{7}, m_{8})$ to vectors $\vec{v} = (v_{0}, v_{1}, v_{2})$ .	`Ascend` `GPU`
`mindsponge.common.rots_scale`	Scaling of rotation matrixs.	`Ascend` `GPU` `CPU`
`mindsponge.common.rots_to_tensor`	Translate rots represented by vectors to tensor, inverse of 'rots_from_tensor'.	`Ascend` `GPU`
`mindsponge.common.vecs_cross_vecs`	Cross product of vectors $v_{1} = (x_{1}, x_{2}, x_{3})$ and $v_{2} = (y_{1}, y_{2}, y_{3})$ .	`Ascend` `GPU`
`mindsponge.common.vecs_dot_vecs`	Dot product of vectors $v_{1} = (x_{1}, x_{2}, x_{3})$ and $v_{2} = (y_{1}, y_{2}, y_{3})$ .	`Ascend` `GPU`
`mindsponge.common.vecs_expand_dims`	Add an extra dimension to the input v at the given axis.	`Ascend` `GPU`
`mindsponge.common.vecs_from_tensor`	Get vectors from the last axis of input tensor.	`Ascend` `GPU`
`mindsponge.common.vecs_robust_norm`	Calculate the l2-norm of a vector.	`Ascend` `GPU` `CPU`
`mindsponge.common.vecs_robust_normalize`	Use l2-norm normalization vectors	`Ascend` `GPU` `CPU`
`mindsponge.common.vecs_scale`	Scale the vector.	`Ascend` `GPU` `CPU`
`mindsponge.common.vecs_sub`	Subtract two vectors.	`Ascend` `GPU` `CPU`
`mindsponge.common.vecs_to_tensor`	Converts 'v' to tensor with last dim shape 3, inverse of 'vecs_from_tensor'.	`Ascend` `GPU`