fix some doc strings

Author: KiddoZhu

CMakeLists.txt

@@ -1,31 +1,34 @@
-if (WIN32)
-    cmake_minimum_required(VERSION 3.8)
+if (WIN32 OR APPLE)
+    set(SKIP_TOOLCHAIN TRUE)
+endif()
+
+if (SKIP_TOOLCHAIN)
+    # skip toolchain so that code insight can work properly
+    cmake_minimum_required(VERSION 3.0)
     project(graphvite LANGUAGES CXX)
-else ()
-    cmake_minimum_required(VERSION 3.12)
-    project(graphvite LANGUAGES CXX CUDA)
-endif ()
+    include_directories(include)
+    add_subdirectory(src)
+    return()
+endif()
+
+cmake_minimum_required(VERSION 3.12)
+project(graphvite LANGUAGES CXX CUDA)
 
 list(APPEND CMAKE_MODULE_PATH ${PROJECT_SOURCE_DIR}/cmake)
 include(FindCUDA)
 
-if (WIN32)
-    find_package(Glog)
-    find_package(GFlags)
-else ()
-    find_package(Glog)
-    find_package(GFlags)
-    find_package(PythonLibsNew REQUIRED)
-    find_package(pybind11 REQUIRED)
-endif ()
+find_package(Glog)
+find_package(GFlags)
+find_package(PythonLibsNew REQUIRED)
+find_package(pybind11 REQUIRED)
 
 # CUDA_ROOT & CUDA_ARCH
 get_filename_component(CUDA_ROOT ${CMAKE_CUDA_COMPILER} DIRECTORY)
 get_filename_component(CUDA_ROOT ${CUDA_ROOT} DIRECTORY)
 if (NOT ALL_ARCH)
     cuda_select_nvcc_arch_flags(CUDA_ARCH Auto)
 else()
-    cuda_select_nvcc_arch_flags(CUDA_ARCH 5.2 6.0 7.0)
+    cuda_select_nvcc_arch_flags(CUDA_ARCH 3.5 5.0 6.0 7.0)
 endif()
 string(REPLACE ";" " " CUDA_ARCH "${CUDA_ARCH}")
 
@@ -57,7 +60,9 @@ if (NOT GLOG_FOUND)
     include_directories(${PROJECT_BINARY_DIR}/glog/include)
     link_directories(${PROJECT_BINARY_DIR}/glog/.libs)
 else()
+    get_filename_component(GLOG_LIBRARY_DIR ${GLOG_LIBRARY} DIRECTORY)
     include_directories(${GLOG_INCLUDE_DIR})
+    link_directories(${GLOG_LIBRARY_DIR})
 endif ()
 
 # gflags
@@ -77,7 +82,9 @@ if (NOT GFLAGS_FOUND)
     include_directories(${PROJECT_BINARY_DIR}/gflags/include)
     link_directories(${PROJECT_BINARY_DIR}/gflags/lib)
 else()
+    get_filename_component(GFLAGS_LIBRARY_DIR ${GFLAGS_LIBRARY} DIRECTORY)
     include_directories(${GFLAGS_INCLUDE_DIR})
+    link_directories(${GFLAGS_LIBRARY_DIR})
 endif()
 
 # faiss

config/word_graph/line_wikipedia.yaml

@@ -16,7 +16,7 @@ build:
     type: SGD
     lr: 0.025
     weight_decay: 0.005
-  num_partition: 0
+  num_partition: auto
   num_negative: 1
   batch_size: 100000
   episode_size: 1000

include/bind.h

@@ -357,7 +357,7 @@ class pyKNNGraph : public py::class_<graphvite::KNNGraph<Index>, graphvite::Grap
                 vectors (2D array_like): vector list
                 num_neighbor (int, optional): number of neighbors for each node
                 perplexity (int, optional): perplexity for the neighborhood of each node
-                normalization (bool, optional): normalize the input vectors or not
+                vector_normalization (bool, optional): normalize the input vectors or not
                 delimiters (str, optional): string of delimiter characters
                 comment (str, optional): prefix of comment strings
             )");
@@ -563,10 +563,10 @@ class pyKnowledgeGraphSolver : public py::class_<graphvite::KnowledgeGraphSolver
             )");
 
         def("train", &KnowledgeGraphSolver::train, py::no_gil(),
-            py::arg("model") = "RotatE", py::arg("num_epoch") = 2000, py::arg("resume") = false, py::arg("margin") = 24,
+            py::arg("model") = "RotatE", py::arg("num_epoch") = 2000, py::arg("resume") = false, py::arg("margin") = 12,
             py::arg("l3_regularization") = 2e-3, py::arg("sample_batch_size") = 2000, py::arg("positive_reuse") = 1,
             py::arg("adversarial_temperature") = 2, py::arg("log_frequency") = 100,
-            "train(model='RotatE', num_epoch=2000, resume=False, margin=24, l3_regularization=2e-3, "
+            "train(model='RotatE', num_epoch=2000, resume=False, margin=12, l3_regularization=2e-3, "
                   "sample_batch_size=2000, positive_reuse=1, adversarial_temperature=2, log_frequency=100)"
             R"(
             Train knowledge graph embeddings.
@@ -681,7 +681,7 @@ class pyVisualizationSolver : public py::class_<graphvite::VisualizationSolver<d
                 resume (bool, optional): resume training from learned embeddings or not
                 sample_batch_size (int, optional): batch size of samples in samplers
                 positive_reuse (int, optional): times of reusing positive samples
-                negative_sample_epoxnent (float, optional): exponent of degrees in negative sampling
+                negative_sample_exponent (float, optional): exponent of degrees in negative sampling
                 negative_weight (float, optional): weight for each negative sample
                 log_frequency (int, optional): log every log_frequency batches
             )");

python/graphvite/__init__.py

@@ -24,11 +24,11 @@
 
 from . import util
 
-package_path = os.path.dirname(os.path.realpath(__file__))
+package_path = os.path.dirname(__file__)
 candidate_paths = [
-    os.path.join(package_path, "lib"),
-    os.path.join(package_path, "../../lib"),
-    os.path.join(package_path, "../../build/lib")
+    os.path.realpath(os.path.join(package_path, "lib")),
+    os.path.realpath(os.path.join(package_path, "../../lib")),
+    os.path.realpath(os.path.join(package_path, "../../build/lib"))
 ]
 lib_file = imp.find_module("libgraphvite", candidate_paths)[1]
 lib_path = os.path.dirname(lib_file)

python/graphvite/cmd.py

@@ -147,7 +147,6 @@ def baseline_main(args):
             file = os.path.join(path, file)
             match = True
             for keyword in args.keywords:
-                # print("(^|_)%s(_|$)" % keyword)
                 result = re.search("[/\_.]%s[/\_.]" % keyword, file)
                 if not result:
                     match = False