update notebook

usc-isi-i2 · May 1, 2017 · 2b4bd44 · 2b4bd44
1 parent ddcebfc
commit 2b4bd44
Show file tree

Hide file tree

Showing 2 changed files with 13 additions and 6 deletions.
diff --git a/docs/notebook_basic.ipynb b/docs/notebook_basic.ipynb
@@ -29,7 +29,7 @@
     "```\n",
     "git clone https://github.com/usc-isi-i2/rltk.git\n",
     "```\n",
-    "2. Create virtual environment (conda should be installed) and install dependencies.\n",
+    "2. Create virtual environment and install dependencies ([Conda](https://github.com/conda/conda) should be installed).\n",
     "```\n",
     "conda-env create .\n",
     "source activate rltk_env\n",

diff --git a/docs/notebook_linkage.ipynb b/docs/notebook_linkage.ipynb
@@ -9,7 +9,7 @@
    "source": [
     "# Record Linkage\n",
     "\n",
-    "Here, let me show you how to do link the actors name from Princeton University Art Museum (PUAM) to Getty Union List of Artist Names (ULAN)."
+    "Here, let me show you how to link the actor names from Princeton University Art Museum (PUAM) to Getty Union List of Artist Names (ULAN)."
    ]
   },
   {
@@ -40,10 +40,10 @@
    "source": [
     "## Prepare data\n",
     "\n",
-    "First and the most important step is preparing data. Except the two candidate datasets (format in json_line/csv/text), you need manually mark some postive and negative pairs of these two datasets. Here, [labeled_puam.jsonl](../examples/puam/labeled_100.jsonl) is a 100 lines labeled paris.\n",
+    "First and the most important step is preparing data. Besides of the two candidate datasets (format in json_line/csv/text), you need manually mark some postive and negative pairs of these two datasets. Here, [labeled_puam.jsonl](../examples/puam/labeled_100.jsonl) is a 100 lines labeled paris.\n",
     "\n",
     "## Get file iterator of datasets\n",
-    "Candidate sets will be streamed as FileIterator in RLTK."
+    "Candidate sets should be streamed as FileIterator in RLTK."
    ]
   },
   {
@@ -167,7 +167,7 @@
     "editable": true
    },
    "source": [
-    "For testing purpose, I create a [blocking_100.jsonl](../examples/puam/blocking_100.jsonl) file which contains first 100 lines of object from `blocking.jsonl`.\n",
+    "For testing purpose, I pick out the first 100 lines of object from the output file `blocking.jsonl` and named it to [blocking_100.jsonl](../examples/puam/blocking_100.jsonl).\n",
     "\n",
     "## Compute vectors and make prediction\n",
     "\n",
@@ -202,6 +202,13 @@
    "source": [
     "tk.predict(model, feature_path='feature.jsonl', predict_output_path='predicted.jsonl')"
    ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Finally, you get the the predicted result of linkage in [predicted.jsonl](../examples/puam/predicted.jsonl)."
+   ]
   }
  ],
  "metadata": {
@@ -224,5 +231,5 @@
   }
  },
  "nbformat": 4,
- "nbformat_minor": 2
+ "nbformat_minor": 0
 }