learning3d 0.1.0__py3-none-any.whl → 0.2.0__py3-none-any.whl

learning3d/losses/cuda/emd_torch/pkg/include/cuda/emd.cuh

@@ -1,347 +0,0 @@
-#ifndef EMD_CUH_
-#define EMD_CUH_
-
-#include "cuda_helper.h"
-
-template<typename T>
-__global__ void approxmatch(const int b, const int n, const int m, const T * __restrict__ xyz1, const T * __restrict__ xyz2, T * __restrict__ match, T * temp){
-	T * remainL=temp+blockIdx.x*(n+m)*2, * remainR=temp+blockIdx.x*(n+m)*2+n,*ratioL=temp+blockIdx.x*(n+m)*2+n+m,*ratioR=temp+blockIdx.x*(n+m)*2+n+m+n;
-	T multiL,multiR;
-	if (n>=m){
-		multiL=1;
-		multiR=n/m;
-	}else{
-		multiL=m/n;
-		multiR=1;
-	}
-	const int Block=1024;
-	__shared__ T buf[Block*4];
-	for (int i=blockIdx.x;i<b;i+=gridDim.x){
-		for (int j=threadIdx.x;j<n*m;j+=blockDim.x)
-			match[i*n*m+j]=0;
-		for (int j=threadIdx.x;j<n;j+=blockDim.x)
-			remainL[j]=multiL;
-		for (int j=threadIdx.x;j<m;j+=blockDim.x)
-			remainR[j]=multiR;
-		__syncthreads();
-		for (int j=7;j>=-2;j--){
-			T level=-powf(4.0f,j);
-			if (j==-2){
-				level=0;
-			}
-			for (int k0=0;k0<n;k0+=blockDim.x){
-				int k=k0+threadIdx.x;
-				T x1=0,y1=0,z1=0;
-				if (k<n){
-					x1=xyz1[i*n*3+k*3+0];
-					y1=xyz1[i*n*3+k*3+1];
-					z1=xyz1[i*n*3+k*3+2];
-				}
-				T suml=T(1e-9f);
-				for (int l0=0;l0<m;l0+=Block){
-					int lend=min(m,l0+Block)-l0;
-					for (int l=threadIdx.x;l<lend;l+=blockDim.x){
-						T x2=xyz2[i*m*3+l0*3+l*3+0];
-						T y2=xyz2[i*m*3+l0*3+l*3+1];
-						T z2=xyz2[i*m*3+l0*3+l*3+2];
-						buf[l*4+0]=x2;
-						buf[l*4+1]=y2;
-						buf[l*4+2]=z2;
-						buf[l*4+3]=remainR[l0+l];
-					}
-					__syncthreads();
-					for (int l=0;l<lend;l++){
-						T x2=buf[l*4+0];
-						T y2=buf[l*4+1];
-						T z2=buf[l*4+2];
-						T d=level*((x2-x1)*(x2-x1)+(y2-y1)*(y2-y1)+(z2-z1)*(z2-z1));
-						T w=__expf(d)*buf[l*4+3];
-						suml+=w;
-					}
-					__syncthreads();
-				}
-				if (k<n)
-					ratioL[k]=remainL[k]/suml;
-			}
-			/*for (int k=threadIdx.x;k<n;k+=gridDim.x){
-				T x1=xyz1[i*n*3+k*3+0];
-				T y1=xyz1[i*n*3+k*3+1];
-				T z1=xyz1[i*n*3+k*3+2];
-				T suml=1e-9f;
-				for (int l=0;l<m;l++){
-					T x2=xyz2[i*m*3+l*3+0];
-					T y2=xyz2[i*m*3+l*3+1];
-					T z2=xyz2[i*m*3+l*3+2];
-					T w=expf(level*((x2-x1)*(x2-x1)+(y2-y1)*(y2-y1)+(z2-z1)*(z2-z1)))*remainR[l];
-					suml+=w;
-				}
-				ratioL[k]=remainL[k]/suml;
-			}*/
-			__syncthreads();
-			for (int l0=0;l0<m;l0+=blockDim.x){
-				int l=l0+threadIdx.x;
-				T x2=0,y2=0,z2=0;
-				if (l<m){
-					x2=xyz2[i*m*3+l*3+0];
-					y2=xyz2[i*m*3+l*3+1];
-					z2=xyz2[i*m*3+l*3+2];
-				}
-				T sumr=0;
-				for (int k0=0;k0<n;k0+=Block){
-					int kend=min(n,k0+Block)-k0;
-					for (int k=threadIdx.x;k<kend;k+=blockDim.x){
-						buf[k*4+0]=xyz1[i*n*3+k0*3+k*3+0];
-						buf[k*4+1]=xyz1[i*n*3+k0*3+k*3+1];
-						buf[k*4+2]=xyz1[i*n*3+k0*3+k*3+2];
-						buf[k*4+3]=ratioL[k0+k];
-					}
-					__syncthreads();
-					for (int k=0;k<kend;k++){
-						T x1=buf[k*4+0];
-						T y1=buf[k*4+1];
-						T z1=buf[k*4+2];
-						T w=__expf(level*((x2-x1)*(x2-x1)+(y2-y1)*(y2-y1)+(z2-z1)*(z2-z1)))*buf[k*4+3];
-						sumr+=w;
-					}
-					__syncthreads();
-				}
-				if (l<m){
-					sumr*=remainR[l];
-					T consumption=fminf(remainR[l]/(sumr+1e-9f),1.0f);
-					ratioR[l]=consumption*remainR[l];
-					remainR[l]=fmaxf(0.0f,remainR[l]-sumr);
-				}
-			}
-			/*for (int l=threadIdx.x;l<m;l+=blockDim.x){
-				T x2=xyz2[i*m*3+l*3+0];
-				T y2=xyz2[i*m*3+l*3+1];
-				T z2=xyz2[i*m*3+l*3+2];
-				T sumr=0;
-				for (int k=0;k<n;k++){
-					T x1=xyz1[i*n*3+k*3+0];
-					T y1=xyz1[i*n*3+k*3+1];
-					T z1=xyz1[i*n*3+k*3+2];
-					T w=expf(level*((x2-x1)*(x2-x1)+(y2-y1)*(y2-y1)+(z2-z1)*(z2-z1)))*ratioL[k];
-					sumr+=w;
-				}
-				sumr*=remainR[l];
-				T consumption=fminf(remainR[l]/(sumr+1e-9f),1.0f);
-				ratioR[l]=consumption*remainR[l];
-				remainR[l]=fmaxf(0.0f,remainR[l]-sumr);
-			}*/
-			__syncthreads();
-			for (int k0=0;k0<n;k0+=blockDim.x){
-				int k=k0+threadIdx.x;
-				T x1=0,y1=0,z1=0;
-				if (k<n){
-					x1=xyz1[i*n*3+k*3+0];
-					y1=xyz1[i*n*3+k*3+1];
-					z1=xyz1[i*n*3+k*3+2];
-				}
-				T suml=0;
-				for (int l0=0;l0<m;l0+=Block){
-					int lend=min(m,l0+Block)-l0;
-					for (int l=threadIdx.x;l<lend;l+=blockDim.x){
-						buf[l*4+0]=xyz2[i*m*3+l0*3+l*3+0];
-						buf[l*4+1]=xyz2[i*m*3+l0*3+l*3+1];
-						buf[l*4+2]=xyz2[i*m*3+l0*3+l*3+2];
-						buf[l*4+3]=ratioR[l0+l];
-					}
-					__syncthreads();
-					T rl=ratioL[k];
-					if (k<n){
-						for (int l=0;l<lend;l++){
-							T x2=buf[l*4+0];
-							T y2=buf[l*4+1];
-							T z2=buf[l*4+2];
-							T w=__expf(level*((x2-x1)*(x2-x1)+(y2-y1)*(y2-y1)+(z2-z1)*(z2-z1)))*rl*buf[l*4+3];
-							match[i*n*m+(l0+l)*n+k]+=w;
-							suml+=w;
-						}
-					}
-					__syncthreads();
-				}
-				if (k<n)
-					remainL[k]=fmaxf(0.0f,remainL[k]-suml);
-			}
-			/*for (int k=threadIdx.x;k<n;k+=blockDim.x){
-				T x1=xyz1[i*n*3+k*3+0];
-				T y1=xyz1[i*n*3+k*3+1];
-				T z1=xyz1[i*n*3+k*3+2];
-				T suml=0;
-				for (int l=0;l<m;l++){
-					T x2=xyz2[i*m*3+l*3+0];
-					T y2=xyz2[i*m*3+l*3+1];
-					T z2=xyz2[i*m*3+l*3+2];
-					T w=expf(level*((x2-x1)*(x2-x1)+(y2-y1)*(y2-y1)+(z2-z1)*(z2-z1)))*ratioL[k]*ratioR[l];
-					match[i*n*m+l*n+k]+=w;
-					suml+=w;
-				}
-				remainL[k]=fmaxf(0.0f,remainL[k]-suml);
-			}*/
-			__syncthreads();
-		}
-	}
-}
-
-void approxmatchLauncher(const int b, const int n, const int m, const at::Tensor xyz1, const at::Tensor xyz2, at::Tensor match, at::Tensor temp){
-	AT_DISPATCH_FLOATING_TYPES(match.type(), "approxmatch", ([&] {
-		approxmatch
-			<<<32, 512>>>(
-			b, n, m, 
-			xyz1.data<scalar_t>(),
-			xyz2.data<scalar_t>(),
-			match.data<scalar_t>(),
-			temp.data<scalar_t>());
-	}));
-	cudaDeviceSynchronize();
-	CUDA_CHECK(cudaGetLastError())
-}
-
-template<typename T>
-__global__ void matchcost(const int b, const int n, const int m, const T * __restrict__ xyz1, const T * __restrict__ xyz2, const T * __restrict__ match, T * __restrict__ out){
-	__shared__ T allsum[512];
-	const int Block=1024;
-	__shared__ T buf[Block*3];
-	for (int i=blockIdx.x;i<b;i+=gridDim.x){
-		T subsum=0;
-		for (int k0=0;k0<n;k0+=blockDim.x){
-			int k=k0+threadIdx.x;
-			T x1=0,y1=0,z1=0;
-			if (k<n){
-				x1=xyz1[i*n*3+k*3+0];
-				y1=xyz1[i*n*3+k*3+1];
-				z1=xyz1[i*n*3+k*3+2];
-			}
-			for (int l0=0;l0<m;l0+=Block){
-				int lend=min(m,l0+Block)-l0;
-				for (int l=threadIdx.x;l<lend*3;l+=blockDim.x)
-					buf[l]=xyz2[i*m*3+l0*3+l];
-				__syncthreads();
-				if (k<n){
-					for (int l=0;l<lend;l++){
-						T x2=buf[l*3+0];
-						T y2=buf[l*3+1];
-						T z2=buf[l*3+2];
-						T d=sqrtf((x2-x1)*(x2-x1)+(y2-y1)*(y2-y1)+(z2-z1)*(z2-z1));
-						subsum+=d*match[i*n*m+(l0+l)*n+k];
-					}
-				}
-				__syncthreads();
-			}
-		}
-		allsum[threadIdx.x]=subsum;
-		for (int j=1;j<blockDim.x;j<<=1){
-			__syncthreads();
-			if ((threadIdx.x&j)==0 && threadIdx.x+j<blockDim.x){
-				allsum[threadIdx.x]+=allsum[threadIdx.x+j];
-			}
-		}
-		if (threadIdx.x==0)
-			out[i]=allsum[0];
-		__syncthreads();
-	}
-}
-
-void matchcostLauncher(const int b, const int n, const int m, const at::Tensor xyz1, const at::Tensor xyz2, const at::Tensor match, at::Tensor out){
-	AT_DISPATCH_FLOATING_TYPES(xyz1.type(), "matchcost", ([&] {
-		matchcost<<<32, 512>>>(
-			b, n, m,
-			xyz1.data<scalar_t>(),
-			xyz2.data<scalar_t>(),
-			match.data<scalar_t>(),
-			out.data<scalar_t>());
-	}));
-	CUDA_CHECK(cudaGetLastError())
-}
-
-template <typename T>
-__global__ void matchcostgrad2(const int b, const int n, const int m,const T * __restrict__ xyz1, const T * __restrict__ xyz2, const T * __restrict__ match, T * __restrict__ grad2){
-	__shared__ T sum_grad[256*3];
-	for (int i=blockIdx.x;i<b;i+=gridDim.x){
-		int kbeg=m*blockIdx.y/gridDim.y;
-		int kend=m*(blockIdx.y+1)/gridDim.y;
-		for (int k=kbeg;k<kend;k++){
-			T x2=xyz2[(i*m+k)*3+0];
-			T y2=xyz2[(i*m+k)*3+1];
-			T z2=xyz2[(i*m+k)*3+2];
-			T subsumx=0,subsumy=0,subsumz=0;
-			for (int j=threadIdx.x;j<n;j+=blockDim.x){
-				T x1=x2-xyz1[(i*n+j)*3+0];
-				T y1=y2-xyz1[(i*n+j)*3+1];
-				T z1=z2-xyz1[(i*n+j)*3+2];
-				T d=match[i*n*m+k*n+j]*rsqrtf(fmaxf(x1*x1+y1*y1+z1*z1,1e-20f));
-				subsumx+=x1*d;
-				subsumy+=y1*d;
-				subsumz+=z1*d;
-			}
-			sum_grad[threadIdx.x*3+0]=subsumx;
-			sum_grad[threadIdx.x*3+1]=subsumy;
-			sum_grad[threadIdx.x*3+2]=subsumz;
-			for (int j=1;j<blockDim.x;j<<=1){
-				__syncthreads();
-				int j1=threadIdx.x;
-				int j2=threadIdx.x+j;
-				if ((j1&j)==0 && j2<blockDim.x){
-					sum_grad[j1*3+0]+=sum_grad[j2*3+0];
-					sum_grad[j1*3+1]+=sum_grad[j2*3+1];
-					sum_grad[j1*3+2]+=sum_grad[j2*3+2];
-				}
-			}
-			if (threadIdx.x==0){
-				grad2[(i*m+k)*3+0]=sum_grad[0];
-				grad2[(i*m+k)*3+1]=sum_grad[1];
-				grad2[(i*m+k)*3+2]=sum_grad[2];
-			}
-			__syncthreads();
-		}
-	}
-}
-
-template <typename T>
-__global__ void matchcostgrad1(const int b, const int n, const int m, const T * __restrict__ xyz1, const T * __restrict__ xyz2, const T * __restrict__ match, T * __restrict__ grad1){
-	for (int i=blockIdx.x;i<b;i+=gridDim.x){
-		for (int l=threadIdx.x;l<n;l+=blockDim.x){
-			T x1=xyz1[i*n*3+l*3+0];
-			T y1=xyz1[i*n*3+l*3+1];
-			T z1=xyz1[i*n*3+l*3+2];
-			T dx=0,dy=0,dz=0;
-			for (int k=0;k<m;k++){
-				T x2=xyz2[i*m*3+k*3+0];
-				T y2=xyz2[i*m*3+k*3+1];
-				T z2=xyz2[i*m*3+k*3+2];
-				T d=match[i*n*m+k*n+l]*rsqrtf(fmaxf((x1-x2)*(x1-x2)+(y1-y2)*(y1-y2)+(z1-z2)*(z1-z2),1e-20f));
-				dx+=(x1-x2)*d;
-				dy+=(y1-y2)*d;
-				dz+=(z1-z2)*d;
-			}
-			grad1[i*n*3+l*3+0]=dx;
-			grad1[i*n*3+l*3+1]=dy;
-			grad1[i*n*3+l*3+2]=dz;
-		}
-	}
-}
-
-void matchcostgradLauncher(const int b, const int n, const int m, const at::Tensor xyz1, const at::Tensor xyz2, const at::Tensor match, at::Tensor grad1, at::Tensor grad2){
-	AT_DISPATCH_FLOATING_TYPES(xyz1.type(), "matchcostgrad1", ([&] {
-		matchcostgrad1<<<32,512>>>(
-			b, n, m, 
-			xyz1.data<scalar_t>(),
-			xyz2.data<scalar_t>(),
-			match.data<scalar_t>(),
-			grad1.data<scalar_t>());
-	}));
-	CUDA_CHECK(cudaGetLastError())
-
-	AT_DISPATCH_FLOATING_TYPES(xyz1.type(), "matchcostgrad2", ([&] {
-		matchcostgrad2<<<dim3(32,32),256>>>(
-			b, n, m,
-			xyz1.data<scalar_t>(),
-			xyz2.data<scalar_t>(),
-			match.data<scalar_t>(),
-			grad2.data<scalar_t>());
-	}));
-	CUDA_CHECK(cudaGetLastError())
-}
-
-#endif

learning3d/losses/cuda/emd_torch/pkg/include/cuda_helper.h

@@ -1,18 +0,0 @@
-#ifndef CUDA_HELPER_H_
-#define CUDA_HELPER_H_
-
-#define CUDA_CHECK(err) \
-	if (cudaSuccess != err) \
-	{ \
-		fprintf(stderr, "CUDA kernel failed: %s (%s:%d)\n", \
-			cudaGetErrorString(err),  __FILE__, __LINE__); \
-		std::exit(-1); \
-	}
-
-#define CHECK_CUDA(x) AT_CHECK(x.type().is_cuda(), \
-	#x " must be a CUDA tensor")
-#define CHECK_CONTIGUOUS(x) AT_CHECK(x.is_contiguous(), \
-	#x " must be contiguous")
-#define CHECK_INPUT(x) CHECK_CUDA(x); CHECK_CONTIGUOUS(x)
-
-#endif

learning3d/losses/cuda/emd_torch/pkg/include/emd.h

@@ -1,54 +0,0 @@
-#ifndef EMD_H_
-#define EMD_H_
-
-#include <torch/extension.h>
-#include <vector>
-
-#include "cuda_helper.h"
-
-
-std::vector<at::Tensor> emd_forward_cuda(
-	at::Tensor xyz1,
-	at::Tensor xyz2);
-
-std::vector<at::Tensor> emd_backward_cuda(
-	at::Tensor xyz1, 
-	at::Tensor xyz2, 
-	at::Tensor match);
-
-// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
-// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
-// CALL FUNCTION IMPLEMENTATIONS
-// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
-// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
-
-std::vector<at::Tensor> emd_forward(
-	at::Tensor xyz1, 
-	at::Tensor xyz2)
-{
-	CHECK_INPUT(xyz1);
-	CHECK_INPUT(xyz2);
-
-	return emd_forward_cuda(xyz1, xyz2);
-}
-
-std::vector<at::Tensor> emd_backward(
-	at::Tensor xyz1, 
-	at::Tensor xyz2,
-	at::Tensor match)
-{
-	CHECK_INPUT(xyz1);
-	CHECK_INPUT(xyz2);
-	CHECK_INPUT(match);
-
-	return emd_backward_cuda(xyz1, xyz2, match);
-}
-
-PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) {
-	m.def("emd_forward", &emd_forward, "Compute Earth Mover's Distance");
-	m.def("emd_backward", &emd_backward, "Compute Gradients for Earth Mover's Distance");
-}
-
-
-
-#endif

learning3d/losses/cuda/emd_torch/pkg/layer/__init__.py

@@ -1 +0,0 @@
-from .emd_loss_layer import EMDLoss

learning3d/losses/cuda/emd_torch/pkg/layer/emd_loss_layer.py

@@ -1,40 +0,0 @@
-import torch
-import torch.nn as nn
-
-import _emd_ext._emd as emd
-
-
-class EMDFunction(torch.autograd.Function):
-	@staticmethod
-	def forward(self, xyz1, xyz2):
-		cost, match = emd.emd_forward(xyz1, xyz2)
-		self.save_for_backward(xyz1, xyz2, match)
-		return cost
-
-
-	@staticmethod
-	def backward(self, grad_output):
-		xyz1, xyz2, match = self.saved_tensors
-		grad_xyz1, grad_xyz2 = emd.emd_backward(xyz1, xyz2, match)
-		return grad_xyz1, grad_xyz2
-
-
-
-
-class EMDLoss(nn.Module):
-	'''
-	Computes the (approximate) Earth Mover's Distance between two point sets. 
-
-	IMPLEMENTATION LIMITATIONS:
-	- Double tensors must have <=11 dimensions
-	- Float tensors must have <=23 dimensions
-	This is due to the use of CUDA shared memory in the computation. This shared memory is limited by the hardware to 48kB.
-	'''
-
-	def __init__(self):
-		super(EMDLoss, self).__init__()
-
-	def forward(self, xyz1, xyz2):
-
-		assert xyz1.shape[-1] == xyz2.shape[-1], 'Both point sets must have the same dimensionality'
-		return EMDFunction.apply(xyz1, xyz2)

learning3d/losses/cuda/emd_torch/pkg/src/cuda/emd.cu

@@ -1,70 +0,0 @@
-#include <ATen/ATen.h>
-
-#include <vector>
-
-#include "cuda/emd.cuh"
-
-
-std::vector<at::Tensor> emd_forward_cuda(
-	at::Tensor xyz1, // B x N1 x D
-	at::Tensor xyz2) // B x N2 x D
-{
-	// Some useful values
-	const int batch_size = xyz1.size(0);
-	const int num_pts_1 = xyz1.size(1);
-	const int num_pts_2 = xyz2.size(1);
-
-	// Allocate necessary data structures
-	at::Tensor match = at::zeros({batch_size, num_pts_1, num_pts_2}, 
-		xyz1.options());
-	at::Tensor cost = at::zeros({batch_size}, xyz1.options());
-	at::Tensor temp = at::zeros({batch_size, 2 * (num_pts_1 + num_pts_2)}, 
-		xyz1.options());
-
-	// Find the approximate matching
-	approxmatchLauncher(
-		batch_size, num_pts_1, num_pts_2,
-		xyz1,
-		xyz2, 
-		match,
-		temp
-	);
-
-	// Compute the matching cost
-	matchcostLauncher(
-		batch_size, num_pts_1, num_pts_2, 
-		xyz1,
-		xyz2, 
-		match,
-		cost
-	);
-
-	return {cost, match};
-}
-
-std::vector<at::Tensor> emd_backward_cuda(
-	at::Tensor xyz1, 
-	at::Tensor xyz2, 
-	at::Tensor match)
-{
-	// Some useful values
-	const int batch_size = xyz1.size(0);
-	const int num_pts_1 = xyz1.size(1);
-	const int num_pts_2 = xyz2.size(1);
-
-	// Allocate necessary data structures
-	at::Tensor grad_xyz1 = at::zeros_like(xyz1);
-	at::Tensor grad_xyz2 = at::zeros_like(xyz2);
-
-	// Compute the gradient with respect to the two inputs (xyz1 and xyz2)
-	matchcostgradLauncher(
-		batch_size, num_pts_1, num_pts_2,
-		xyz1,
-		xyz2,
-		match,
-		grad_xyz1,
-		grad_xyz2
-	);
-
-	return {grad_xyz1, grad_xyz2};
-}

learning3d/losses/cuda/emd_torch/pkg/src/emd.cpp

@@ -1 +0,0 @@
-#include "emd.h"

learning3d/losses/cuda/emd_torch/setup.py

@@ -1,29 +0,0 @@
-from setuptools import setup
-from torch.utils.cpp_extension import BuildExtension, CUDAExtension
-
-
-setup(
-	name='PyTorch EMD',
-	version='0.0',
-	author='Vinit Sarode',
-	author_email='vinitsarode5@gmail.com', 
-	description='A PyTorch module for the earth mover\'s distance loss', 
-	ext_package='_emd_ext',
-	ext_modules=[
-		CUDAExtension(
-			name='_emd', 
-			sources=[
-				'pkg/src/emd.cpp',
-				'pkg/src/cuda/emd.cu',
-			],
-			include_dirs=['pkg/include'],
-		),
-	],
-	packages=[
-		'emd', 
-	],
-	package_dir={
-		'emd' : 'pkg/layer'
-	},
-	cmdclass={'build_ext': BuildExtension}, 
-)

learning3d/utils/__init__.py

@@ -1,4 +0,0 @@
-from .svd import SVDHead
-from .transformer import Transformer, Identity
-from .ppfnet_util import angle_difference, square_distance, index_points, farthest_point_sample, query_ball_point, sample_and_group, sample_and_group_multi
-from .pointconv_util import PointConvDensitySetAbstraction

learning3d-0.1.0.dist-info/RECORD

@@ -1,80 +0,0 @@
-learning3d/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-learning3d/data_utils/__init__.py,sha256=iYAVh0FThnVlG42QIgmDYrC3NGVYuzKX8s1oRqAI1YU,261
-learning3d/data_utils/dataloaders.py,sha256=QlRd7c7ir19bMliZXq9PF-iNrQ_BtNTonpBCpmjYiFU,14675
-learning3d/data_utils/user_data.py,sha256=ADDGeCUCr6TcXhcxvAFncIeLO71xoRHYi4H418ktvQs,4828
-learning3d/examples/test_dcp.py,sha256=EYXgCz0yh2-bjJE9s8XEUY96LfUE9_hZcdMffyrW3x0,5783
-learning3d/examples/test_deepgmr.py,sha256=f9nv7Y5se9_iZ3UGasTUDnErBZh3FX0fRNY4IqeqjCk,5774
-learning3d/examples/test_masknet.py,sha256=-xiGaaG9SN_u1POh51sM358QFsyx9QoYIhaNWtBNCXY,6558
-learning3d/examples/test_masknet2.py,sha256=ABNDRa9R6Sp9S_H1euj2bx3LnLt6ONKRNvp5bat0jAQ,7012
-learning3d/examples/test_pcn.py,sha256=rvKsqbCRmj4CLU1xWI-VzUh63t89QY6sIHi-39uhzmM,4519
-learning3d/examples/test_pcrnet.py,sha256=JTq7rQN5cFxucVD-0gQ6t0K0ASuX4b_sNI7uxgUKsdg,4519
-learning3d/examples/test_pnlk.py,sha256=WhZfD9s-YUg-nne_JCOSIcs0akJJV30G19QBkMVK2gc,4609
-learning3d/examples/test_pointconv.py,sha256=CA-Z3KLGXtqkCWIUi1IIlpaHCyTMMd5QuNyi37D8Iz8,4826
-learning3d/examples/test_pointnet.py,sha256=Cd0h-x4UcuiV-hDVMeriOYN75-uc_ZQ34mn2ikLEIbg,4477
-learning3d/examples/test_prnet.py,sha256=F5Ju6BafuDocPa3qMa9ZDASaE9zP_5vgp1hgRkpIvso,5078
-learning3d/examples/test_rpmnet.py,sha256=dKORbcOuil2qlUrY5VdHtaUUOGxDc8Paj7FVYmcHqEs,4629
-learning3d/examples/train_PointNetLK.py,sha256=MnYJuU1ThTLRSuylqXZp6Eakeggk2Xd99BJuTyVIvxI,8950
-learning3d/examples/train_dcp.py,sha256=QcyFKKCKwUH3RDUQU7-eET7Q3tIc5hLGlBBrMwv4ans,9688
-learning3d/examples/train_deepgmr.py,sha256=I9GBE78u5tVrKo0zmRDjrTk0Lq-R8TRH27mEKYJU5us,9574
-learning3d/examples/train_masknet.py,sha256=MRlySX4Llfp_0Eu8jwkHg3FdLlWLtlB4tVUNFbSsBbM,9066
-learning3d/examples/train_pcn.py,sha256=QKJLY31Hcx8w2K4XhmY3U3zho5vFePEWr7xHo4fAUvY,7719
-learning3d/examples/train_pcrnet.py,sha256=pUBYK1gtDFhBBon1UBF2yB5-2cGhILkntmplOYRBqiQ,8375
-learning3d/examples/train_pointconv.py,sha256=Vn_WghOFYOOiL-U7yJQ3TB7osYTLbwEQ1WVWTvBhPVs,9116
-learning3d/examples/train_pointnet.py,sha256=bpPMt0QtUoJlVgoMoPX191sO25VCk14pIKdBTCPS62I,9017
-learning3d/examples/train_prnet.py,sha256=aruTiJB8WCad0cbsu5XdfhMSybdibWSGRPilbzi6xWI,8550
-learning3d/examples/train_rpmnet.py,sha256=Q_kB7UnRdcIpFk_w47oNcuhIsmHLLij4dS7vz2GzTIY,8707
-learning3d/losses/__init__.py,sha256=zjjZeA_NvAhZlxiYBbtgjEsvMyLFhFWXlZioitrlGWw,425
-learning3d/losses/chamfer_distance.py,sha256=UTZ6x5cGwL3L5hJZOWoC35gTzcKh1S0yCg8vGuGXU1w,2121
-learning3d/losses/classification.py,sha256=QgDHC5VgSga6BiuD4Ee70t6vvchlE97BY0KExevhdgk,374
-learning3d/losses/correspondence_loss.py,sha256=Tcq2o5eLY7j50pibAuH0vBcUTjwZ-wHNzGZD4H6mAe0,583
-learning3d/losses/emd.py,sha256=DqP77dN6lPkpGGgBz10oO6YNYxt889weYbVYj6bZFUM,389
-learning3d/losses/frobenius_norm.py,sha256=IuKr0DT9aPBlc5fjIy6lJ082yOh9F8xiNoXF6FvWZtY,682
-learning3d/losses/rmse_features.py,sha256=_KMqIWqH9-lH2P6YSeGfSOIbP7plUAwWWBh2Cu7cpXA,453
-learning3d/losses/cuda/chamfer_distance/__init__.py,sha256=5nz6Ui4PVKru4HLGovFue3M-eumQlovMYP6Qfk3O6SE,46
-learning3d/losses/cuda/chamfer_distance/chamfer_distance.cpp,sha256=rcknAz6oYPhlTVjGnAga2c4bMz2rz_UU9ch7zrcmWoc,6260
-learning3d/losses/cuda/chamfer_distance/chamfer_distance.cu,sha256=7-UQ8QVcSQ1tbN7xwDNQQfpbG-S5b7KuSv0nCBYlhI8,5070
-learning3d/losses/cuda/chamfer_distance/chamfer_distance.py,sha256=vzF4GdkgdD5Rr3F6pk1u1kF1LpiFKpLTZcd-H1IVu3s,1926
-learning3d/losses/cuda/emd_torch/setup.py,sha256=AUvCT6EI2fOcCbyzrGscvCKu5gLhjofmZ-PfNBY2dNw,574
-learning3d/losses/cuda/emd_torch/pkg/emd_loss_layer.py,sha256=FMfOoT8VF05Ns6_BaF6tquT5B9T7man0WtKJ9lPRfk8,1116
-learning3d/losses/cuda/emd_torch/pkg/include/cuda_helper.h,sha256=8uVFZEr7YWVMPqHqfu_5lLUHCDjgwE0bJc-MUhmbLRk,469
-learning3d/losses/cuda/emd_torch/pkg/include/emd.h,sha256=5TF1_V_OFywM8ftegQGw0YpdgPAzg6UolJ4OOFG5rVM,1215
-learning3d/losses/cuda/emd_torch/pkg/include/cuda/emd.cuh,sha256=XliqHIePtL73Kdh0eDlRLalNmP3KqzwAEcgWysAOnNs,9826
-learning3d/losses/cuda/emd_torch/pkg/layer/__init__.py,sha256=aDpfP0iZyg3Uu-ppa33u2imCsmlvKSf5S8QhLnGfwyI,35
-learning3d/losses/cuda/emd_torch/pkg/layer/emd_loss_layer.py,sha256=yCEEfafLZ1Ia_BCrE7mcnDRDaaEj6je3Rj8cnQ_Zrts,1019
-learning3d/losses/cuda/emd_torch/pkg/src/emd.cpp,sha256=lQ4q2XO5K2CldYmnRJnGhKTR9hVRFTwO305bHT3Cauo,17
-learning3d/losses/cuda/emd_torch/pkg/src/cuda/emd.cu,sha256=DJXALRWyiVDzaKWsD2tQnEXrIT9GpIldkvIJ9fZMGbI,1462
-learning3d/models/__init__.py,sha256=WgAx7FlaUbijjl3qjQCvwyrbdhxvKHV7HbqBlq47ux8,473
-learning3d/models/classifier.py,sha256=_LUNXbLrpKNXmCkO2R1mz64dbwfrze7f_4SYT1Z6SYo,1205
-learning3d/models/dcp.py,sha256=LZFgtk9f9f9s3QvX65nFXGgC33yGIZuy4XjviwH8OGE,3377
-learning3d/models/deepgmr.py,sha256=vIxOQrZjvOCHLElJCjZ8EcZ-vm0-v71IKsPGuSF-elE,5298
-learning3d/models/dgcnn.py,sha256=Bt-dP2NwpOy4AcWrspXfVV1EKL-ZQElYUp2uaWNvE_Q,3057
-learning3d/models/masknet.py,sha256=ElMF3b-JgYmgwSEf1taGQvhA7Xy7_MiHEofzc03VCd8,2705
-learning3d/models/masknet2.py,sha256=6lgukurfzUOY-6xdCpMljOYFtvADLSczAXJzRC3Jkh4,9063
-learning3d/models/pcn.py,sha256=FvpjLR6t3kFQ1I4Fhpbsaj_P8Ml6S912x36JAZ1dUKs,5346
-learning3d/models/pcrnet.py,sha256=6C6iM3XkDNdgihtPIdy09RgFD2KKDCnDzLvFfp6X-Tg,2755
-learning3d/models/pointconv.py,sha256=lJ3_3uslE29lO3roZiE5vxr5971AWV5ExeVTzbEl858,5151
-learning3d/models/pointnet.py,sha256=qgSWLJ4N5Y7ObAwKiJH29Pcl67jm3sfqbXqi3tQbUQg,3238
-learning3d/models/pointnetlk.py,sha256=Zl66LjDX1vLdZRgCdY2oQJnpWpqPEx6BH8GbcVCsw68,5805
-learning3d/models/pooling.py,sha256=vOzJMambeG7kf1pnox5p5FE2CVH2iMDGU_DgWRw15WQ,412
-learning3d/models/ppfnet.py,sha256=aBzWvtNHFo-eu1kWoZmPI9xJOFHyxYHjdapb6cN2Aso,2894
-learning3d/models/prnet.py,sha256=KSTTMEUgEvE2PX59Mqas48LQUYyh8O6PwOHvP8zEYHU,18499
-learning3d/models/rpmnet.py,sha256=eMVqJ6BalY96TSB8VFXjCJIA15J0XAB3BEpMB-6CMdM,11517
-learning3d/models/segmentation.py,sha256=CjlINj5M0Y6C-CejrieIu9ZkuwEoCFNjq_hr5SX9umU,1166
-learning3d/ops/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-learning3d/ops/data_utils.py,sha256=qW8FOQWgcHjuuiSaAayJ3nKJnDW_GDv3ujE-uFvWsPE,1764
-learning3d/ops/invmat.py,sha256=m1Mm2mQNn4KgQv54w-ek0xLkZuKnfDlF2ooylXeXvAw,4154
-learning3d/ops/quaternion.py,sha256=D00IL6VHitYy4CINFAilD0t0kyBjc_JHfKuMrJGJ1Cw,6793
-learning3d/ops/se3.py,sha256=x6oLbQzLOXiM0xDJsVUCUE1liZ_TaJzkkHQvIyjqCqI,3957
-learning3d/ops/sinc.py,sha256=A_Ffu07RXGx3daZn4zOGdnW10_l06cmhFdAiU4NKhcw,5228
-learning3d/ops/so3.py,sha256=b0tX5nHyF2Qtp8V0ejGKaPaHJQ_G38ifQ7gSJzRU1ts,5166
-learning3d/ops/transform_functions.py,sha256=hvNjZO-uJodsGYtQwtAtDxtQ6uBpA7Lv9t-_yAg6wxo,12806
-learning3d/utils/__init__.py,sha256=lwMtW0H4LEZJuoVCi7qmfaWdOVrL2YmP_YP-ZfUoSlc,286
-learning3d/utils/pointconv_util.py,sha256=kJxGztai7X15YsGuorMOc50SPtj_k1yfkP4XCTzIWdM,14331
-learning3d/utils/ppfnet_util.py,sha256=HEoxkgUBlawKZLWspfQm3caWUyAMIrW-ECtStNYbe2Y,7989
-learning3d/utils/svd.py,sha256=yCYQt2SKqeIzCBnBEr_8xFR79m4fIoNVFnp77epn1dM,1936
-learning3d/utils/transformer.py,sha256=UDgJvnh7ekWyijaAn-a3ckeFeMxlK_chXzWlhAGDiPM,8974
-learning3d-0.1.0.dist-info/LICENSE,sha256=3qY3_NeQIvalbLlsHFtOfuUKjs_U2k6u7rf6YVx6ac0,1098
-learning3d-0.1.0.dist-info/METADATA,sha256=kKzkNAYiiVf_t_gv-_OGGr3C1QZWQ5BCKh5sPZUbLts,15813
-learning3d-0.1.0.dist-info/WHEEL,sha256=GJ7t_kWBFywbagK5eo9IoUwLW6oyOeTKmQ-9iHFVNxQ,92
-learning3d-0.1.0.dist-info/top_level.txt,sha256=nTmYW8NhbNV1_15DGNpl_OvvSFtQP98sy3qrrHr0eLo,11
-learning3d-0.1.0.dist-info/RECORD,,

learning3d-0.1.0.dist-info/top_level.txt

@@ -1 +0,0 @@
-learning3d