data inspection notebooks

notebooks/data_analytics_qas.ipynb

@@ -0,0 +1,352 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "UsageError: Line magic function `%` not found.\n"
+     ]
+    }
+   ],
+   "source": [
+    "import os\n",
+    "import h5py\n",
+    "import json\n",
+    "import numpy as np\n",
+    "import sys\n",
+    "import glob\n",
+    "import typing\n",
+    "import copy\n",
+    "from typing import List, Dict, Union\n",
+    "import matplotlib.pyplot as plt\n",
+    "import pickle as pkl\n",
+    "% matplotlib inline\n",
+    "sys.path.append(\"../\") # Append visdialch path"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from collections import defaultdict\n",
+    "unique_dialog_ques = defaultdict()\n",
+    "unique_dialog_ans = defaultdict()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "path_train_data = \"../../data/visdial_1.0_train.json\"\n",
+    "dense_annotations_jsonpath = \"../../data/visdial_1.0_train_dense_annotations.json\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "data_train = json.load(open(path_train_data))\n",
+    "\n",
+    "questions = data_train['data']['questions']\n",
+    "len(questions)\n",
+    "\n",
+    "answers = data_train['data']['answers']\n",
+    "len(answers)\n",
+    "\n",
+    "dialogs = data_train['data']['dialogs']\n",
+    "\n",
+    "unique_dialog_ques = defaultdict()\n",
+    "unique_dialog_ans = defaultdict()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "123287\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(len(dialogs))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "for dial_index in range(len(dialogs)):\n",
+    "    dial_for_image = dialogs[dial_index]['dialog']\n",
+    "    for round_index in range(len(dial_for_image)):\n",
+    "        ques_key = questions[dial_for_image[round_index]['question']]\n",
+    "        ans_key = answers[dial_for_image[round_index]['answer']]        \n",
+    "        unique_dialog_ques[ques_key] = unique_dialog_ques.get(ques_key, 0) + 1\n",
+    "        unique_dialog_ans[ans_key] = unique_dialog_ans.get(ans_key, 0) + 1"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "337527\n",
+      "OrderedDict([('no', 262225), ('yes', 236115), ('white', 19680), ('black', 11818), ('brown', 8401), (\"can't tell\", 8195), (\"i can't tell\", 6519), ('2', 5012), ('yes it is', 4798), ('1', 4500), ('0', 4312), ('not that i can see', 3581), ('nope', 3514), ('blue', 3389), ('not really', 3336), ('3', 3274), ('red', 3216), ('day', 3152), ('i think so', 3046), ('green', 2946)])\n"
+     ]
+    }
+   ],
+   "source": [
+    "from collections import Counter, OrderedDict\n",
+    "\n",
+    "# unique_dialog_ans = dict(unique_dialog_ans)\n",
+    "unique_ans = Counter(unique_dialog_ans)\n",
+    "most_common_ans = OrderedDict(unique_ans.most_common(20))\n",
+    "\n",
+    "print(len(unique_ans))\n",
+    "print(most_common_ans)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "376082\n",
+      "is it sunny -- 13466\n",
+      "is it daytime -- 10241\n",
+      "is the photo in color -- 7279\n",
+      "can you see the sky -- 5082\n",
+      "are there any people -- 4894\n",
+      "what color is it -- 4314\n",
+      "any people -- 3914\n",
+      "is it day or night -- 3654\n",
+      "is it sunny out -- 3618\n",
+      "is this in color -- 3049\n",
+      "do you see any people -- 3035\n",
+      "what color are the walls -- 2953\n",
+      "can you see any people -- 2946\n",
+      "is it a sunny day -- 2734\n",
+      "any trees -- 2666\n",
+      "how old is the man -- 2657\n",
+      "are there trees -- 2655\n",
+      "what is he wearing -- 2614\n",
+      "what color is the table -- 2570\n",
+      "what color is his hair -- 2560\n"
+     ]
+    }
+   ],
+   "source": [
+    "unique_ques = Counter(unique_dialog_ques)\n",
+    "most_common_ques = OrderedDict(unique_ques.most_common(20))\n",
+    "\n",
+    "print(len(unique_ques))\n",
+    "# print(most_common_ques)\n",
+    "\n",
+    "for item in most_common_ques.items():\n",
+    "    print(item[0], \"--\", item[1])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 31,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Collecting pandas\n",
+      "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/73/9b/52e228545d14f14bb2a1622e225f38463c8726645165e1cb7dde95bfe6d4/pandas-0.25.1-cp36-cp36m-manylinux1_x86_64.whl (10.5MB)\n",
+      "\u001b[K    100% |████████████████████████████████| 10.5MB 2.2MB/s eta 0:00:01\n",
+      "\u001b[?25hRequirement already satisfied: pytz>=2017.2 in /home/shubham/anaconda3/envs/visdialch/lib/python3.6/site-packages (from pandas) (2017.3)\n",
+      "Requirement already satisfied: numpy>=1.13.3 in /home/shubham/anaconda3/envs/visdialch/lib/python3.6/site-packages (from pandas) (1.15.4)\n",
+      "Requirement already satisfied: python-dateutil>=2.6.1 in /home/shubham/anaconda3/envs/visdialch/lib/python3.6/site-packages (from pandas) (2.8.0)\n",
+      "Requirement already satisfied: six>=1.5 in /home/shubham/anaconda3/envs/visdialch/lib/python3.6/site-packages (from python-dateutil>=2.6.1->pandas) (1.12.0)\n",
+      "Installing collected packages: pandas\n",
+      "Successfully installed pandas-0.25.1\n"
+     ]
+    }
+   ],
+   "source": [
+    "%!pip install pandas"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "ques_df = pd.DataFrame.from_dict(unique_ques, orient='index')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>0</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <td>is this a child or adult</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>male or female</td>\n",
+       "      <td>838</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>are they inside or outside</td>\n",
+       "      <td>381</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>are they laying on the floor</td>\n",
+       "      <td>3</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>is the floor carpeted or wooden</td>\n",
+       "      <td>2</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>how big is the gate</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>any of them have tusks</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>would you go in this building</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>is he doing a flip in the air</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>is he wearing any kind of gear</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>376082 rows × 1 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                                   0\n",
+       "is this a child or adult           1\n",
+       "male or female                   838\n",
+       "are they inside or outside       381\n",
+       "are they laying on the floor       3\n",
+       "is the floor carpeted or wooden    2\n",
+       "...                              ...\n",
+       "how big is the gate                1\n",
+       "any of them have tusks             1\n",
+       "would you go in this building      1\n",
+       "is he doing a flip in the air      1\n",
+       "is he wearing any kind of gear     1\n",
+       "\n",
+       "[376082 rows x 1 columns]"
+      ]
+     },
+     "execution_count": 10,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "ques_df"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.6.9"
+  },
+  "pycharm": {
+   "stem_cell": {
+    "cell_type": "raw",
+    "source": [],
+    "metadata": {
+     "collapsed": false
+    }
+   }
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}