Merge branch 'master' into evaluation

Author: kinkunchan

semb/methods/struc2vec/.gitignore

@@ -0,0 +1 @@
+random_walks.txt

semb/methods/struc2vec/algorithms_distances.py

@@ -607,7 +607,7 @@ def generate_distances_network(workers):
     logging.info('Creating distance network...')
 
     os.system("rm "+returnPickle().replace(' ', '\ ')+"weights_distances-layer-*.pickle")
-    # print('!!!!!')
+
     print("rm "+returnPickle().replace(' ', '\ ')+"weights_distances-layer-*.pickle")
     with ProcessPoolExecutor(max_workers=1) as executor:
         job = executor.submit(generate_distances_network_part1,workers)

semb/methods/struc2vec/method.py

@@ -8,6 +8,8 @@
 from .utils import *
 from .graph import *
 from semb.methods import BaseMethod
+from .utils import *
+from .graph import *
 
 """
 Driver for learning Struc2Vec embedding

semb/methods/struc2vec/pickles/.gitignore

@@ -0,0 +1 @@
+*.pickle

semb/methods/struc2vec/pickles/.gitkeep

@@ -218,8 +218,3 @@ def get_representations(graph, rep_method, verbose = True):
 # 	rep_method = RepMethod(max_layer = 2) #Learn representations with xNetMF.  Can adjust parameters (e.g. as in REGAL)
 # 	representations = get_representations(graph, rep_method)
 # 	print(representations.shape)
-
-
-
-
-

semb/methods/xnetmf/internal/xnetmf.py

@@ -1,4 +1,5 @@
 {
+
  "cells": [
   {
    "cell_type": "markdown",
@@ -339,6 +340,76 @@
     "perform_clustering(emb, dict_labels)"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Struc2Vec"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "output_type": "stream",
+     "name": "stdout",
+     "text": "!!!!!\nrm /Users/jackiezhang/Projects/Research/StrucEmbeddingLibrary/semb/methods/struc2vec/pickles/weights_distances-layer-*.pickle\n<gensim.models.keyedvectors.Word2VecKeyedVectors object at 0x7ff73958afd0>\n"
+    }
+   ],
+   "source": [
+    "from semb.methods import load as load_method\n",
+    "from semb.datasets import load as load_dataset\n",
+    "\n",
+    "Struc2VecMethod = load_method(\"struc2vec\")\n",
+    "AirportDataProvider = load_dataset(\"airports\")\n",
+    "airport_datasets = AirportDataProvider().get_datasets()\n",
+    "brazil_airport_graph = AirportDataProvider().load_dataset(airport_datasets[0])\n",
+    "\n",
+    "struc2vec = Struc2VecMethod(brazil_airport_graph, num_walks=10, walk_length=80, window_size=10, dim=128, opt1=True, opt2=True, opt3=True, until_layer=2)\n",
+    "struc2vec.train()\n",
+    "\n",
+    "print(struc2vec.get_embeddings())"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# XNetMF"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "output_type": "stream",
+     "name": "stdout",
+     "text": "max degree:  80\ngot k hop neighbors in time:  0.027379989624023438\ngot degree sequences in time:  0.011226892471313477\ncomputed representation in time:  0.009149789810180664\n[[-1.37704743e-04  1.27010683e-03 -2.03434516e-04 ... -1.49213045e-22\n  -3.41994083e-23  1.43444650e-22]\n [-9.65898443e-05  1.66742756e-04 -2.45790879e-04 ... -1.66882048e-22\n  -4.09051438e-23  1.70831618e-22]\n [-1.58249968e-05 -1.83503745e-05  1.23592685e-04 ... -1.73237629e-22\n  -4.80606517e-23  1.70797436e-22]\n ...\n [-6.44032227e-05 -1.28126308e-04 -7.84104267e-06 ... -3.57732969e-22\n  -1.28058809e-22  2.18954869e-22]\n [-5.36337990e-05 -1.31538894e-04  5.44271448e-06 ... -3.59625795e-22\n  -1.28893017e-22  2.24111482e-22]\n [ 4.40716560e-04  7.88162017e-06 -1.71904982e-04 ... -3.50170074e-22\n  -1.17795556e-22  2.16827953e-22]]\n"
+    }
+   ],
+   "source": [
+    "from semb.methods import load as load_method\n",
+    "from semb.datasets import load as load_dataset\n",
+    "\n",
+    "Struc2VecMethod = load_method(\"xnetmf\")\n",
+    "AirportDataProvider = load_dataset(\"airports\")\n",
+    "airport_datasets = AirportDataProvider().get_datasets()\n",
+    "brazil_airport_graph = AirportDataProvider().load_dataset(airport_datasets[0])\n",
+    "\n",
+    "struc2vec = Struc2VecMethod(brazil_airport_graph, dim=128, max_layer=2, discount=0.1, gamma=1)\n",
+    "struc2vec.train()\n",
+    "\n",
+    "print(struc2vec.get_embeddings())"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,