demo.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Loading word embeddings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import io\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_vec(emb_path, nmax=50000):\n",
    "    vectors = []\n",
    "    word2id = {}\n",
    "    with io.open(emb_path, 'r', encoding='utf-8', newline='\\n', errors='ignore') as f:\n",
    "        next(f)\n",
    "        for i, line in enumerate(f):\n",
    "            word, vect = line.rstrip().split(' ', 1)\n",
    "            vect = np.fromstring(vect, sep=' ')\n",
    "            assert word not in word2id, 'word found twice'\n",
    "            vectors.append(vect)\n",
    "            word2id[word] = len(word2id)\n",
    "            if len(word2id) == nmax:\n",
    "                break\n",
    "    id2word = {v: k for k, v in word2id.items()}\n",
    "    embeddings = np.vstack(vectors)\n",
    "    return embeddings, id2word, word2id"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "path = \"/Users/avij1/Desktop/imp_shit/fasttext/MUSE/dumped/debug/375u9teo6w\"\n",
    "p2 = \"/Users/avij1/Desktop/imp_shit/fasttext/MUSE/dumped/debug/375u9teo6w/vectors-hi.txt\"\n",
    "# path = '/Users/avij1/Desktop/imp_shit/fasttext/MUSE/dumped/debug/w1c3c03p4w'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['params.pkl',\n",
       " 'vectors-en.txt',\n",
       " 'vectors-hi.txt',\n",
       " 'best_mapping.pth',\n",
       " 'train.log']"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "os.listdir(path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "src_path = os.path.join(path,'vectors-hi.txt')\n",
    "tgt_path = os.path.join(path,'vectors-en.txt')\n",
    "nmax = 50000  # maximum number of word embeddings to load\n",
    "\n",
    "src_embeddings, src_id2word, src_word2id = load_vec(src_path, nmax)\n",
    "tgt_embeddings, tgt_id2word, tgt_word2id = load_vec(tgt_path, nmax)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Get nearest neighbors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_nn(word, src_emb, src_id2word, tgt_emb, tgt_id2word, K=5):\n",
    "    print(\"Nearest neighbors of \\\"%s\\\":\" % word)\n",
    "    word2id = {v: k for k, v in src_id2word.items()}\n",
    "    word_emb = src_emb[word2id[word]]\n",
    "    scores = (tgt_emb / np.linalg.norm(tgt_emb, 2, 1)[:, None]).dot(word_emb / np.linalg.norm(word_emb))\n",
    "    k_best = scores.argsort()[-K:][::-1]\n",
    "    for i, idx in enumerate(k_best):\n",
    "        print('%.4f - %s' % (scores[idx], tgt_id2word[idx]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Nearest neighbors of \"cat\":\n",
      "1.0000 - cat\n",
      "0.4227 - dog\n",
      "0.4182 - hat\n",
      "0.4164 - लेखसभी\n",
      "0.3989 - needing\n"
     ]
    }
   ],
   "source": [
    "# printing nearest neighbors in the source space\n",
    "src_word = 'cat'\n",
    "get_nn(src_word, src_embeddings, src_id2word, src_embeddings, src_id2word, K=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Nearest neighbors of \"cat\":\n",
      "0.3292 - rss\n",
      "0.3192 - stubs\n",
      "0.3112 - ref\n",
      "0.3111 - wp\n",
      "0.3064 - articles\n"
     ]
    }
   ],
   "source": [
    "# printing nearest neighbors in the target space\n",
    "src_word = 'cat'\n",
    "get_nn(src_word, src_embeddings, src_id2word, tgt_embeddings, tgt_id2word, K=5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Visualize multilingual embeddings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Variance explained: 0.10\n"
     ]
    }
   ],
   "source": [
    "from sklearn.decomposition import PCA\n",
    "pca = PCA(n_components=2, whiten=True)  # TSNE(n_components=2, n_iter=3000, verbose=2)\n",
    "pca.fit(np.vstack([src_embeddings, tgt_embeddings]))\n",
    "print('Variance explained: %.2f' % pca.explained_variance_ratio_.sum())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "\n",
    "def plot_similar_word(src_words, src_word2id, src_emb, tgt_words, tgt_word2id, tgt_emb, pca):\n",
    "\n",
    "    Y = []\n",
    "    word_labels = []\n",
    "    for sw in src_words:\n",
    "        Y.append(src_emb[src_word2id[sw]])\n",
    "        word_labels.append(sw)\n",
    "    for tw in tgt_words:\n",
    "        Y.append(tgt_emb[tgt_word2id[tw]])\n",
    "        word_labels.append(tw)\n",
    "\n",
    "    # find tsne coords for 2 dimensions\n",
    "    Y = pca.transform(Y)\n",
    "    x_coords = Y[:, 0]\n",
    "    y_coords = Y[:, 1]\n",
    "\n",
    "    # display scatter plot\n",
    "    plt.figure(figsize=(10, 8), dpi=80)\n",
    "    plt.scatter(x_coords, y_coords, marker='x')\n",
    "\n",
    "    for k, (label, x, y) in enumerate(zip(word_labels, x_coords, y_coords)):\n",
    "        color = 'blue' if k < len(src_words) else 'red'  # src words in blue / tgt words in red\n",
    "        plt.annotate(label, xy=(x, y), xytext=(0, 0), textcoords='offset points', fontsize=19,\n",
    "                     color=color, weight='bold')\n",
    "\n",
    "    plt.xlim(x_coords.min() - 0.2, x_coords.max() + 0.2)\n",
    "    plt.ylim(y_coords.min() - 0.2, y_coords.max() + 0.2)\n",
    "    plt.title('Visualization of the multilingual word embedding space')\n",
    "\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda2/envs/fasttext/lib/python3.6/site-packages/matplotlib/backends/backend_agg.py:211: RuntimeWarning: Glyph 2325 missing from current font.\n",
      "  font.set_text(s, 0.0, flags=flags)\n",
      "/anaconda2/envs/fasttext/lib/python3.6/site-packages/matplotlib/backends/backend_agg.py:211: RuntimeWarning: Glyph 2361 missing from current font.\n",
      "  font.set_text(s, 0.0, flags=flags)\n",
      "/anaconda2/envs/fasttext/lib/python3.6/site-packages/matplotlib/backends/backend_agg.py:211: RuntimeWarning: Glyph 2366 missing from current font.\n",
      "  font.set_text(s, 0.0, flags=flags)\n",
      "/anaconda2/envs/fasttext/lib/python3.6/site-packages/matplotlib/backends/backend_agg.py:211: RuntimeWarning: Glyph 2305 missing from current font.\n",
      "  font.set_text(s, 0.0, flags=flags)\n",
      "/anaconda2/envs/fasttext/lib/python3.6/site-packages/matplotlib/backends/backend_agg.py:180: RuntimeWarning: Glyph 2325 missing from current font.\n",
      "  font.set_text(s, 0, flags=flags)\n",
      "/anaconda2/envs/fasttext/lib/python3.6/site-packages/matplotlib/backends/backend_agg.py:180: RuntimeWarning: Glyph 2361 missing from current font.\n",
      "  font.set_text(s, 0, flags=flags)\n",
      "/anaconda2/envs/fasttext/lib/python3.6/site-packages/matplotlib/backends/backend_agg.py:180: RuntimeWarning: Glyph 2366 missing from current font.\n",
      "  font.set_text(s, 0, flags=flags)\n",
      "/anaconda2/envs/fasttext/lib/python3.6/site-packages/matplotlib/backends/backend_agg.py:180: RuntimeWarning: Glyph 2305 missing from current font.\n",
      "  font.set_text(s, 0, flags=flags)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x640 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# get 5 random input words\n",
    "src_words = ['कहाँ' ] \n",
    "tgt_words = ['where']\n",
    "\n",
    "# assert words in dictionaries\n",
    "for sw in src_words:\n",
    "    assert sw in src_word2id, '\"%s\" not in source dictionary' % sw\n",
    "for tw in tgt_words:\n",
    "    assert tw in tgt_word2id, '\"%s\" not in target dictionary' % sw\n",
    "\n",
    "plot_similar_word(src_words, src_word2id, src_embeddings, tgt_words, tgt_word2id, tgt_embeddings, pca)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "map1 = torch.load(os.path.join(path,'best_mapping.pth'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 0.01308666, -0.00946762,  0.08758792, ..., -0.03236237,\n",
       "         0.05270232,  0.03733799],\n",
       "       [ 0.07407887,  0.02124145, -0.05285997, ..., -0.0092205 ,\n",
       "        -0.07361791,  0.07491986],\n",
       "       [ 0.04116647, -0.03698866,  0.00020679, ..., -0.02390712,\n",
       "         0.08679352, -0.00853956],\n",
       "       ...,\n",
       "       [ 0.00909949, -0.03637621,  0.00812729, ..., -0.02400344,\n",
       "         0.12105906,  0.03065394],\n",
       "       [ 0.01647672,  0.02596837, -0.00890792, ..., -0.0957305 ,\n",
       "         0.05216065,  0.12187852],\n",
       "       [-0.00297859, -0.12286018,  0.00149462, ..., -0.01722991,\n",
       "         0.01299748,  0.00137717]], dtype=float32)"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "map1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(300, 300)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "map1.shape"
   ]
  },
  {
   "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"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}