.

Author: Halvani

Demo_POSNoise.ipynb

@@ -2,27 +2,38 @@
  "cells": [
   {
    "cell_type": "markdown",
-   "id": "22430569",
+   "id": "77dd4a9b",
    "metadata": {},
    "source": [
     "## POSNoise"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
-   "id": "a6e994f3",
+   "execution_count": 1,
+   "id": "381f1f4b",
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[POSNoise] Loading spaCy model 'en_core_web_sm' (disabled: parser, ner)...\n",
+      "[POSNoise] Loaded 'en_core_web_sm'.\n",
+      "The original dataset contains two partitions comprising sockpuppets and non-sockpuppets cases.\n",
+      "The @ # Ø two # Ø # and @@@ #.\n"
+     ]
+    }
+   ],
    "source": [
     "import posnoise\n",
     "from posnoise import POSNoise"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
-   "id": "5f770232",
+   "execution_count": 2,
+   "id": "a54b16a6",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -31,8 +42,8 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
-   "id": "130a1ee7",
+   "execution_count": 4,
+   "id": "6febdb03",
    "metadata": {
     "scrolled": true
    },
@@ -41,13 +52,15 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Mathematics is the most beautiful and most powerful creation of the human spirit.\n",
-      "# is the most @ and most @ # of the @ #.\n"
+      "I love python !\n",
+      "I Ø # !\n"
      ]
     }
    ],
    "source": [
-    "document = \"Mathematics is the most beautiful and most powerful creation of the human spirit.\"\n",
+    "#document = \"Mathematics is the most beautiful and most powerful creation of the human spirit.\"\n",
+    "\n",
+    "document = \"I love python !\"\n",
     "posnoised_doc = posnoise_instance.pos_noise(document)\n",
     "\n",
     "print(document)\n",
@@ -57,7 +70,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "41ceb46c",
+   "id": "69bf62a8",
    "metadata": {},
    "outputs": [],
    "source": []

posnoise/core.py

@@ -296,10 +296,3 @@ def pos_noise(self, text: str) -> str:
                 text = text[: token.idx] + replace_token + text[token.idx + len(token.text) :]
         return text
 
-
-
-posnoise = POSNoise(spacy_model_size=SpacyModelSize.Small, verbose=True)
-document = "The original dataset contains two partitions comprising sockpuppets and non-sockpuppets cases."
-posnoised_doc = posnoise.pos_noise(document)
-print(document)
-print(posnoised_doc)

tests/test_suite.py

@@ -21,9 +21,9 @@ class TestPOSNoise(unittest.TestCase):
 
     def test_posnoise_masking_1_large_model(self):
         posnoise_instance = POSNoise(spacy_model_size=posnoise.SpacyModelSize.Large)
-        document = "Mathematics is the most beautiful and most powerful creation of the human spirit."
+        document = "I love python !"
         posnoised_doc = posnoise_instance.pos_noise(document) 
 
-        assert posnoised_doc == "# is the most @ and most @ # of the @ #."
+        assert posnoised_doc == "I Ø # !"