Removed prints

Author: nicolas-chaulet

README.md

@@ -2,13 +2,13 @@
 Pytorch kernels for pointnet++ like architectures
 
 ## Installation
-Requires torch version 1.0 or higher.
+Requires torch version 1.0 or higher. To install a specific version replace the version number with a given tag.
 ```
-pip install git+https://github.com/nicolas-chaulet/torch-points.git
+pip install git+https://github.com/nicolas-chaulet/torch-points.git#v0.2.3
 ```
 or with poetry:
 ```
-poetry add git+https://github.com/nicolas-chaulet/torch-points.git
+poetry add git+https://github.com/nicolas-chaulet/torch-points.git#v0.2.3
 ```
 
 ## Usage

cpu/src/torch_nearest_neighbors.cpp

@@ -82,15 +82,6 @@ std::pair<at::Tensor, at::Tensor> batch_ball_query(at::Tensor support, at::Tenso
     std::vector<long> support_batch_stl(support_batch.DATA_PTR<long>(),
                                         support_batch.DATA_PTR<long>() + support_batch.numel());
 
-    std::cout << std::endl;
-    for (auto i : query_batch_stl)
-        std::cout << i << " ";
-    std::cout << std::endl;
-
-    for (auto i : support_batch_stl)
-        std::cout << i << " ";
-    std::cout << std::endl;
-
     AT_DISPATCH_ALL_TYPES(query.scalar_type(), "batch_radius_search", [&] {
         std::vector<scalar_t> queries_stl(query.DATA_PTR<scalar_t>(),
                                           query.DATA_PTR<scalar_t>() + query.numel());
@@ -104,9 +95,6 @@ std::pair<at::Tensor, at::Tensor> batch_ball_query(at::Tensor support, at::Tenso
     auto neighbors_dists_ptr = neighbors_dists.data();
     long* neighbors_indices_ptr = neighbors_indices.data();
 
-    for (auto i : neighbors_indices)
-        std::cout << i << " ";
-    std::cout << std::endl;
     if (mode == 0)
     {
         idx =

test/test_ballquerry.py

@@ -88,23 +88,23 @@ def test_simple_cpu(self):
         npt.assert_array_almost_equal(dist2, dist2_answer)
 
     def test_random_cpu(self):
-        a = torch.randn(10, 3).to(torch.float)
-        b = torch.randn(10, 3).to(torch.float)
+        a = torch.randn(100, 3).to(torch.float)
+        b = torch.randn(50, 3).to(torch.float)
         batch_a = torch.tensor([0 for i in range(a.shape[0] // 2)] + [1 for i in range(a.shape[0] // 2, a.shape[0])])
         batch_b = torch.tensor([0 for i in range(b.shape[0] // 2)] + [1 for i in range(b.shape[0] // 2, b.shape[0])])
         R = 1
 
-        idx, dist = ball_query(R, 5, a, b, mode="PARTIAL_DENSE", batch_x=batch_a, batch_y=batch_b)
+        idx, dist = ball_query(R, 15, a, b, mode="PARTIAL_DENSE", batch_x=batch_a, batch_y=batch_b)
         self.assertEqual(idx.shape[0], b.shape[0])
         self.assertEqual(dist.shape[0], b.shape[0])
-        self.assertLessEqual(idx.max().item(), len(batch_b))
+        self.assertLessEqual(idx.max().item(), len(batch_a))
 
         # Comparison to see if we have the same result
         tree = KDTree(a.detach().numpy())
         idx3_sk = tree.query_radius(b.detach().numpy(), r=R)
         i = np.random.randint(len(batch_b))
         for p in idx[i].detach().numpy():
-            if p < len(batch_b):
+            if p < len(batch_a):
                 assert p in idx3_sk[i]