add msc working utils

1 files changed, 287 insertions, 0 deletions

diff --git a/megapixels/notebooks/datasets/vgg_face1/flickr_meta.ipynb b/megapixels/notebooks/datasets/vgg_face1/flickr_meta.ipynb
new file mode 100644
index 00000000..99bbe32e
--- /dev/null
+++ b/megapixels/notebooks/datasets/vgg_face1/flickr_meta.ipynb
@@ -0,0 +1,287 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# VGG Face (V1) Prepare Flickr API"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 46,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%reload_ext autoreload\n",
+    "%autoreload 2\n",
+    "\n",
+    "import os\n",
+    "from os.path import join\n",
+    "from glob import glob, iglob\n",
+    "from pathlib import Path\n",
+    "from tqdm import tqdm_notebook as tqdm\n",
+    "\n",
+    "import pandas as pd\n",
+    "import sys\n",
+    "sys.path.append('/work/megapixels_dev/megapixels/')\n",
+    "from app.utils import file_utils"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Convert annotation files to list of photo IDs"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 47,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "fp_dir_annos = '/data_store/datasets/people/vgg_face/downloads/vgg_face_dataset/files/'\n",
+    "fp_photo_ids = '/data_store/datasets/people/vgg_face/research/photo_ids.csv'"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 50,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "b92b24eac4c84f2f96e32f6eba8d2dc0",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=0, max=2622), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "photo_ids = []\n",
+    "all_photos = []\n",
+    "fp_annos = glob(join(fp_dir_annos, '*.txt'))\n",
+    "for fp_anno in tqdm(fp_annos):\n",
+    "  df_annos = pd.read_csv(fp_anno, delimiter=' ', names=['url', 'a', 'b', 'c', 'd', 'e', 'f', 'g'])\n",
+    "  records = df_annos.to_dict('records')\n",
+    "  for record in records:\n",
+    "    url = record['url']\n",
+    "    all_photos.append(url)\n",
+    "    if 'flickr.com' in url:\n",
+    "      photo_id = Path(url).stem.split('_')[0]\n",
+    "      photo_ids.append({'photo_id': photo_id})"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 51,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "2604849\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(len(all_photos))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "PosixPath('/data_store/datasets/people/vgg_face/research/photo_ids.csv')"
+      ]
+     },
+     "execution_count": 20,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "file_utils.ensure_posixpath(fp_photo_ids)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "pd.DataFrame.from_dict(photo_ids).to_csv(fp_photo_ids, index=False)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "## Convert Flickr API data to filepaths and counts"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 33,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "fp_in_flickr_api = '/data_store_hdd/datasets/people/vgg_face/research/vgg_flickr_api_photo_ids.csv'\n",
+    "fp_out_filepaths = '/data_store_hdd/datasets/people/vgg_face/research/vgg_filepaths.csv'\n",
+    "fp_out_counts = '/data_store_hdd/datasets/people/vgg_face/research/vgg_counts.csv'"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 34,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df = pd.read_csv(fp_in_flickr_api)\n",
+    "records = df.to_dict('records')\n",
+    "\n",
+    "# write filepaths\n",
+    "results = []\n",
+    "for record in records:\n",
+    "  photo_id = record['photo_id']\n",
+    "  obj = {\n",
+    "    'filepath': f'{photo_id}.jpg',\n",
+    "    'nsid': record['nsid'],\n",
+    "    'photo_id': photo_id,\n",
+    "    'url': record['url']\n",
+    "  }\n",
+    "  results.append(obj)\n",
+    "\n",
+    "pd.DataFrame.from_dict(results).to_csv(fp_out_filepaths, index=False)\n",
+    "\n",
+    "# write counts\n",
+    "results = []\n",
+    "nsid_groups = df.groupby('nsid')\n",
+    "for nsid, group in nsid_groups:\n",
+    "  results.append({'nsid': nsid, 'count': len(group)})\n",
+    "\n",
+    "pd.DataFrame.from_dict(results).to_csv(fp_out_counts, index=False)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 35,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "fp = '/data_store/datasets/msc/embassies/embassies_on_flickr.csv'\n",
+    "df = pd.read_csv(fp)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 36,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df_match = df[df['nsid'] == '50747072@N03']"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 39,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "  bureau country          nsid     path_alias       type  \\\n",
+      "0    EUR  Russia  50747072@N03  otkroyameriku  Consulate   \n",
+      "\n",
+      "                                          url                  username  \\\n",
+      "0  http://www.flickr.com/photos/otkroyameriku  Генконсульство США в СПб   \n",
+      "\n",
+      "  verified notes  \n",
+      "0      NaN   NaN   1\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(df_match, len(df_match))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 44,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 45,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "'50747072@N03'"
+      ]
+     },
+     "execution_count": 45,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "match.nsid"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "megapixels",
+   "language": "python",
+   "name": "megapixels"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.6.8"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}