{ "cells": [ { "cell_type": "code", "execution_count": 2, "metadata": { "scrolled": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "#palabras: 43981\n", "# vector: 300\n", "[-3.45608e-02 9.21620e-02 -2.59729e-02 1.02635e-01 -2.01081e-02\n", " -1.74898e-02 6.78648e-02 -8.46215e-02 -4.58716e-02 3.93783e-02\n", " 1.02216e-01 8.79728e-02 1.23162e-01 -5.40405e-02 -2.95337e-02\n", " 6.19999e-02 -2.91148e-02 4.50337e-02 4.58716e-02 -6.49323e-02\n", " 1.92702e-02 2.47162e-02 -4.50337e-02 -2.54493e-02 -2.18885e-02\n", " 5.55067e-03 -4.14729e-02 6.07432e-02 1.97939e-02 1.13632e-02\n", " 3.33040e-02 7.85472e-03 4.54526e-02 1.15203e-02 3.07905e-02\n", " -1.37196e-02 -1.29341e-02 1.57094e-02 2.67060e-02 4.62905e-02\n", " -6.70269e-02 4.26773e-03 8.95438e-03 3.83310e-02 6.87026e-02\n", " -1.36148e-02 1.09757e-01 1.13946e-01 -1.88513e-02 2.57635e-02\n", " -6.61891e-02 -4.16824e-02 -5.08986e-02 -6.70269e-02 1.02216e-01\n", " 8.79728e-02 1.63378e-02 -9.97026e-02 -3.72837e-02 -6.40945e-02\n", " 4.52432e-02 5.82296e-02 -1.18135e-01 -7.62431e-02 -1.52905e-02\n", " -6.70269e-02 -3.66554e-02 1.02635e-01 7.64526e-03 6.36756e-02\n", " -8.79728e-02 -8.43073e-03 9.04864e-02 5.36215e-02 5.55067e-03\n", " 5.86486e-02 1.13946e-01 -2.70202e-02 6.15810e-02 1.85371e-02\n", " -1.87466e-02 2.57635e-02 -7.62431e-02 -5.48783e-02 3.47702e-02\n", " -7.28918e-02 -6.91215e-02 6.24188e-02 8.46215e-02 5.11080e-02\n", " -3.74932e-02 -6.95404e-02 -2.37736e-02 -1.68615e-02 -7.87566e-02\n", " -1.24105e-02 -2.80152e-03 -1.34054e-02 -8.08512e-02 -2.55540e-02\n", " 4.58716e-02 -5.52972e-02 -9.71891e-02 -3.49797e-02 -8.12702e-02\n", " -2.80675e-02 5.65540e-02 3.70743e-02 3.60270e-02 -4.75472e-02\n", " 1.50811e-02 6.91215e-02 -5.82296e-02 4.75472e-02 7.07972e-02\n", " 4.45101e-03 -2.06317e-02 9.58276e-03 2.78581e-02 -3.37229e-02\n", " -8.32600e-03 3.63935e-03 3.37229e-02 -3.62364e-02 2.01081e-02\n", " 2.84864e-02 4.46148e-02 -8.83918e-02 3.39324e-02 3.41418e-02\n", " 2.15743e-02 1.95844e-02 -6.49323e-02 -4.60810e-02 1.69662e-02\n", " -1.35730e-01 1.43480e-02 1.08919e-01 -3.28851e-02 8.53546e-03\n", " -8.25269e-02 -1.28189e-01 6.03242e-02 -1.17821e-02 -1.34054e-02\n", " -4.71283e-02 5.11080e-02 6.19999e-02 7.70810e-02 5.73918e-02\n", " 2.16162e-01 3.07905e-02 -3.95354e-03 -5.83867e-03 -4.73378e-02\n", " -4.33580e-02 -2.17314e-03 -3.45608e-02 -3.83310e-02 -1.61284e-02\n", " -3.51891e-02 3.69172e-03 3.45608e-02 -5.36215e-02 -7.24729e-02\n", " 6.61891e-02 -1.05855e-05 -2.46115e-02 -4.12635e-02 -1.13239e-03\n", " -1.78459e-01 -5.36215e-02 4.90134e-02 -6.15810e-02 9.63512e-02\n", " 8.90201e-03 1.69662e-02 5.82296e-02 -4.52432e-02 1.01378e-01\n", " 1.82229e-02 2.30405e-02 -9.13242e-02 4.69189e-02 -3.58699e-03\n", " -5.78107e-02 8.42026e-02 -2.63919e-02 8.88107e-02 -3.43513e-02\n", " 5.57161e-02 -3.16283e-02 1.72804e-03 -9.32093e-03 1.81182e-02\n", " 3.35135e-02 -1.59189e-02 5.29932e-02 -3.77026e-02 -5.65540e-02\n", " -6.45134e-02 -5.44594e-03 1.00017e-02 6.59796e-03 -1.86838e-01\n", " -2.15743e-02 -6.87026e-02 4.64999e-02 -8.50404e-02 -3.95354e-03\n", " 6.53513e-02 6.54560e-03 -6.70269e-02 -2.97432e-02 1.17821e-02\n", " 9.04864e-02 -2.07365e-02 -1.59189e-02 -4.85945e-02 -6.70269e-02\n", " -5.44594e-02 3.79121e-02 5.61351e-02 2.19932e-02 3.33040e-02\n", " -7.38344e-03 4.60810e-02 -3.24662e-02 1.00017e-02 -7.24729e-02\n", " 6.40945e-02 1.22534e-02 9.88647e-02 -4.41959e-02 9.42566e-02\n", " -1.09442e-02 -1.17297e-02 1.67567e-03 6.70269e-02 -3.41418e-02\n", " 8.50404e-02 -3.95878e-02 -8.54593e-02 2.54493e-02 6.30996e-03\n", " -4.06351e-02 -6.07432e-02 -6.74459e-02 -6.61891e-02 -5.69729e-02\n", " 4.27297e-02 -1.39290e-02 8.90201e-04 1.33007e-02 1.22534e-02\n", " -1.28189e-01 -8.33647e-02 2.07365e-02 -7.16350e-02 -2.60777e-02\n", " 2.70202e-02 -3.69172e-03 -1.18973e-01 1.64425e-02 4.18918e-03\n", " 3.62364e-02 1.43270e-01 2.78843e-03 -1.07871e-02 -5.73918e-02\n", " -4.12635e-02 2.56587e-02 -4.44053e-02 7.83377e-02 1.75946e-02\n", " -1.35101e-02 -6.57178e-03 -1.65892e-01 -1.05149e-01 -8.21080e-02\n", " 7.49864e-02 -3.91689e-02 1.27246e-02 6.80742e-03 -8.16891e-02\n", " 8.12702e-02 -6.74459e-02 3.95878e-02 3.89594e-02 9.73985e-03\n", " 1.21748e-03 -2.74392e-02 -7.28918e-02 -2.37736e-02 1.31436e-02\n", " 2.80675e-02 -7.62431e-02 5.65540e-02 -1.28294e-02 1.03159e-02]\n", "-0.6270861061348114\n", "------\n", "[ 2.48629972e-03 1.03602998e-01 -1.10154012e-02 1.91389710e-01\n", " 1.95145011e-02 -4.42351028e-02 1.31260991e-01 -1.03640392e-01\n", " -4.77660596e-02 2.52930075e-03 2.03253999e-01 1.06496394e-01\n", " 1.50105402e-01 -9.40593034e-02 -7.25241974e-02 1.05188593e-01\n", " -1.10341199e-01 5.07542193e-02 1.01343304e-01 -1.00592703e-01\n", " 1.02477700e-01 9.40558016e-02 -4.03037518e-02 4.42865044e-02\n", " -1.99197493e-02 -1.36298329e-01 -4.05504368e-02 1.35629907e-01\n", " 6.83315992e-02 1.16039999e-03 7.47095942e-02 -3.55320796e-02\n", " 6.16979003e-02 1.45662902e-02 -3.53792980e-02 -7.43421987e-02\n", " 7.97827989e-02 -4.69619967e-03 4.55266982e-02 9.70074981e-02\n", " -6.17273673e-02 5.61733283e-02 5.37279807e-02 1.78262983e-02\n", " 8.26696008e-02 4.50267009e-02 1.79492801e-01 1.02653600e-01\n", " -6.38229996e-02 1.18876696e-01 -1.13736197e-01 -9.12107006e-02\n", " -5.23411110e-02 -1.13187298e-01 1.88197106e-01 3.18039954e-03\n", " -2.60583982e-02 -8.18723962e-02 -4.22860608e-02 -1.28679395e-01\n", " 9.38469917e-03 4.19842973e-02 -7.49462992e-02 -9.68469009e-02\n", " 1.62095018e-02 -7.59299845e-03 -1.06787398e-01 1.86634898e-01\n", " 1.66098792e-02 8.01189989e-02 -1.62463307e-01 4.07012701e-02\n", " 1.25750497e-01 6.85790032e-02 -3.96191292e-02 2.02874601e-01\n", " 2.30435997e-01 -2.57262699e-02 1.31713003e-01 8.27257037e-02\n", " 1.46090053e-03 -4.19912003e-02 -1.37261897e-01 2.95179002e-02\n", " 5.71569987e-02 -5.68445995e-02 -5.40649034e-02 1.58701807e-01\n", " 1.51979893e-01 3.96173969e-02 -4.48729098e-02 -1.46408290e-01\n", " -2.93207690e-02 -7.90690035e-02 -8.59877318e-02 2.58252993e-02\n", " 6.17833845e-02 1.23492992e-02 -1.06209695e-01 3.56860086e-03\n", " 2.64565013e-02 -6.07205480e-02 -1.53453201e-01 -1.01165008e-02\n", " -1.69628605e-01 -6.19447976e-02 6.18040003e-02 5.96592017e-02\n", " 2.72109695e-02 -1.31943405e-01 -7.84282982e-02 1.38064906e-01\n", " -7.18993992e-02 2.92216986e-02 1.37759387e-01 4.34885779e-03\n", " 6.50979951e-03 -1.06247403e-02 -2.95947008e-02 -9.94964018e-02\n", " -5.38920015e-02 -2.52692006e-03 4.37771007e-02 -4.50524315e-02\n", " -4.06170078e-03 1.01609007e-02 1.11337006e-02 -1.43467203e-01\n", " 3.51767987e-02 6.74248040e-02 5.84232993e-02 -1.24641597e-01\n", " -1.43781304e-01 -8.74865949e-02 -3.69205996e-02 -1.47864401e-01\n", " 2.53928006e-02 1.08914770e-01 -4.41775024e-02 3.70637588e-02\n", " -2.26752907e-01 -2.72414982e-01 1.13022298e-01 -9.92479082e-03\n", " -2.40045004e-02 4.00415026e-02 1.18815973e-02 2.33677998e-02\n", " 6.80668503e-02 5.75844944e-02 3.80199999e-01 -4.13227007e-02\n", " -4.06044386e-02 5.47839291e-02 -1.30941510e-01 -9.56598967e-02\n", " 1.01637863e-01 -1.31636202e-01 5.25390357e-03 -4.56473008e-02\n", " -1.27509803e-01 4.98521216e-02 1.17372103e-01 2.56237015e-02\n", " -1.03972897e-01 7.08695203e-02 -1.24916853e-02 -1.09800100e-01\n", " -1.64002981e-02 2.02638097e-02 -2.66817391e-01 -3.76733989e-02\n", " 5.54520786e-02 -1.68561995e-01 1.03186101e-01 3.54492068e-02\n", " 2.39496902e-02 1.07956000e-01 -4.10085283e-02 2.01622993e-01\n", " 1.69351604e-02 4.66159992e-02 -8.43902454e-02 6.67302012e-02\n", " 3.08847111e-02 -5.26102334e-02 4.93347012e-02 -5.19484989e-02\n", " 1.80338889e-01 3.09209898e-03 8.52349997e-02 -6.82791993e-02\n", " -7.08798971e-03 6.31884709e-02 -9.04477984e-02 2.30502337e-04\n", " 3.08357999e-02 -2.30821967e-02 -4.06866819e-02 -6.44785240e-02\n", " -1.07503898e-01 -1.43610295e-02 9.67752039e-02 -1.08308040e-01\n", " -2.14177608e-01 -1.40442297e-01 -6.16200604e-02 -5.65190986e-02\n", " -1.05347000e-01 -4.53240238e-03 1.01605996e-01 -5.48694991e-02\n", " 5.08630276e-03 -1.74762189e-01 -1.20557897e-01 2.22033411e-01\n", " -6.19440004e-02 8.25090148e-03 -1.03669897e-01 -1.25272095e-01\n", " -1.04978204e-01 4.68701124e-04 6.62883967e-02 8.14270973e-02\n", " 8.32284987e-02 3.71085992e-03 -5.85229993e-02 1.34960003e-02\n", " 4.11055014e-02 -8.74304026e-02 1.48490697e-01 4.43476997e-02\n", " 1.19072199e-01 2.11814009e-02 1.02874525e-01 2.31312998e-02\n", " -8.03983957e-03 -7.67771304e-02 7.36141577e-02 -4.61276025e-02\n", " 1.13568693e-01 -6.10831007e-02 8.05009902e-03 -1.00129005e-02\n", " -1.04633033e-01 -8.77860039e-02 -1.27309203e-01 -9.41911936e-02\n", " -5.95523119e-02 -5.94616979e-02 5.57061024e-02 2.56936010e-02\n", " 5.71543016e-02 7.11496994e-02 -2.32088007e-02 -2.36754999e-01\n", " -3.64119038e-02 1.01870010e-02 -6.38095289e-02 7.61482939e-02\n", " 4.56428006e-02 5.29685766e-02 -2.19218001e-01 -2.10008994e-02\n", " 4.30193767e-02 1.34534016e-02 1.58029407e-01 2.97318306e-02\n", " -2.97068991e-02 -3.31229009e-02 -5.59985638e-05 6.63980097e-03\n", " -7.55630061e-03 1.63130105e-01 -1.41035002e-02 -1.15736097e-01\n", " 1.17843226e-01 -2.93476999e-01 -1.29814103e-01 -2.08108008e-01\n", " 2.74393000e-02 -5.96735999e-02 -4.27470990e-02 1.09033413e-01\n", " -1.59349397e-01 8.10720921e-02 -1.05285496e-01 1.19229302e-01\n", " 1.00374497e-01 1.30984843e-01 -4.25655209e-02 -3.71467397e-02\n", " -1.32722005e-01 -5.09150997e-02 5.57378978e-02 4.78919968e-03\n", " -1.19233601e-01 5.14773503e-02 2.40195990e-02 5.90516999e-02]\n" ] } ], "source": [ "# Word embeddings\n", "import gensim\n", "from nltk.data import find\n", "\n", "# Cargar el modelo de embeding pre-entrenados del NLTK\n", "word2vec_sample = str(find('models/word2vec_sample/pruned.word2vec.txt'))\n", "model = gensim.models.KeyedVectors.load_word2vec_format(word2vec_sample, binary=False)\n", "\n", "print(\"#palabras: \", len(model))\n", "\n", "# Tamaño del vector\n", "print (\"# vector: \",len(model['class']))\n", "print (model['class'])\n", "print (sum(model['class']))\n", "a=[]\n", "a.append(model['class'])\n", "a.append(model['school'])\n", "print(\"------\")\n", "print(a[0]+a[1])" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[('soccer', 0.7313547134399414),\n", " ('Football', 0.7124834656715393),\n", " ('basketball', 0.6682467460632324),\n", " ('athletics', 0.6265193223953247),\n", " ('baseball', 0.6161999106407166),\n", " ('sports', 0.5927177667617798),\n", " ('coaches', 0.579153835773468),\n", " ('athletic', 0.5767539739608765),\n", " ('Soccer', 0.5632802248001099),\n", " ('hockey', 0.5496212244033813)]" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "#Buscar n palabras similares a una dada\n", "model.most_similar(positive=['football'], topn = 10)" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "'house'" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "#Palabras que no son de un dominio temático\n", "\n", "model.doesnt_match(['breakfast','house', 'dinner', 'lunch'])" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [], "source": [ "#Mikolov et al. (2013) figured out that word embedding captures much of syntactic and semantic regularities. \n", "#For example, the vector 'King - Man + Woman' is close to 'Queen' and 'Germany - Berlin + Paris' is close to 'France'." ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[('queen', 0.7118193507194519)]" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "model.most_similar(positive=['woman','king'], negative=['man'], topn = 1)" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[('France', 0.7884091734886169)]" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ "\n", "model.most_similar(positive=['Paris','Germany'], negative=['Berlin'], topn = 1)" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0.054031882" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "model.similarity('woman','bad')" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0.12128081" ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "model.similarity('man','bad')" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0.7190051" ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" } ], "source": [ "model.similarity('good','bad')" ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0.22174425" ] }, "execution_count": 11, "metadata": {}, "output_type": "execute_result" } ], "source": [ "model.similarity('handsome','ugly')" ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0.44486597" ] }, "execution_count": 12, "metadata": {}, "output_type": "execute_result" } ], "source": [ "model.similarity('handsome','beautiful')" ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[('queen', 0.7118193507194519),\n", " ('monarch', 0.6189674139022827),\n", " ('princess', 0.5902430415153503),\n", " ('prince', 0.5377321243286133),\n", " ('kings', 0.5236843228340149),\n", " ('queens', 0.5181134939193726),\n", " ('throne', 0.5005807280540466),\n", " ('royal', 0.493820458650589),\n", " ('ruler', 0.49092739820480347),\n", " ('princes', 0.481081485748291)]" ] }, "execution_count": 13, "metadata": {}, "output_type": "execute_result" } ], "source": [ "model.most_similar(positive=['king','woman'],negative=['man'])" ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0.2596795" ] }, "execution_count": 14, "metadata": {}, "output_type": "execute_result" } ], "source": [ "model.similarity('chocolate','salt')" ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0.42128563" ] }, "execution_count": 15, "metadata": {}, "output_type": "execute_result" } ], "source": [ "model.similarity('chocolate','sweet')" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": 16, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "fi\n" ] } ], "source": [ "print (\"fi\")" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "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": "3.9.18" } }, "nbformat": 4, "nbformat_minor": 2 }