math/py-annoy: New port: Approximate Nearest Neighbors in C++

4 files changed, 40 insertions, 0 deletions

diff --git a/math/py-annoy/files/patch-setup.py b/math/py-annoy/files/patch-setup.py
new file mode 100644
index 000000000000..f1bf9c98ae64
--- /dev/null
+++ b/math/py-annoy/files/patch-setup.py
@@ -0,0 +1,10 @@
+--- setup.py.orig	2023-01-12 05:20:26 UTC
++++ setup.py
+@@ -104,6 +104,6 @@ setup(name='annoy',
+           'Programming Language :: Python :: 3.9',
+       ],
+       keywords='nns, approximate nearest neighbor search',
+-      setup_requires=['nose>=1.0'],
++      setup_requires=[],
+       tests_require=['numpy', 'h5py']
+       )
diff --git a/math/py-annoy/files/patch-src_annoymodule.cc b/math/py-annoy/files/patch-src_annoymodule.cc
new file mode 100644
index 000000000000..6c0be8088fde
--- /dev/null
+++ b/math/py-annoy/files/patch-src_annoymodule.cc
@@ -0,0 +1,11 @@
+--- src/annoymodule.cc.orig	2023-01-12 04:57:07 UTC
++++ src/annoymodule.cc
+@@ -179,7 +179,7 @@ py_an_init(py_annoy *self, PyObject *args, PyObject *k
+   int f;
+   static char const * kwlist[] = {"f", "metric", NULL};
+   if (!PyArg_ParseTupleAndKeywords(args, kwargs, "i|s", (char**)kwlist, &f, &metric))
+-    return (int) NULL;
++    return 0;
+   return 0;
+ }
+ 
diff --git a/math/py-annoy/files/test-load.py b/math/py-annoy/files/test-load.py
new file mode 100644
index 000000000000..3eecbf6bb041
--- /dev/null
+++ b/math/py-annoy/files/test-load.py
@@ -0,0 +1,7 @@
+from annoy import AnnoyIndex
+
+f = 40  # Length of item vector that will be indexed
+
+u = AnnoyIndex(f, 'angular')
+u.load('test.ann') # super fast, will just mmap the file
+print(u.get_nns_by_item(0, 1000)) # will find the 1000 nearest neighbors
diff --git a/math/py-annoy/files/test-save.py b/math/py-annoy/files/test-save.py
new file mode 100644
index 000000000000..21a8f50df0c0
--- /dev/null
+++ b/math/py-annoy/files/test-save.py
@@ -0,0 +1,12 @@
+from annoy import AnnoyIndex
+import random
+
+f = 40  # Length of item vector that will be indexed
+
+t = AnnoyIndex(f, 'angular')
+for i in range(1000):
+    v = [random.gauss(0, 1) for z in range(f)]
+    t.add_item(i, v)
+
+t.build(10) # 10 trees
+t.save('test.ann')