{ "cells": [ { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "import pandas as pd\n", "from IPython.display import clear_output, Markdown, Math\n", "import ipywidgets as widgets\n", "import os\n", "import glob\n", "import json\n", "import numpy as np\n", "import itertools\n", "import sklearn.utils as sku\n", "from sklearn.feature_extraction.text import CountVectorizer, TfidfVectorizer, HashingVectorizer\n", "from sklearn.model_selection import train_test_split\n", "from sklearn.preprocessing import MultiLabelBinarizer" ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [], "source": [ "# TODO importing function for sentiment to emoji mapping\n", "def sent2emoji(sentiment_vector: np.array):\n", " return '๐คฆ'\n", "\n", "def emoji2sent(emoji: str):\n", " return np.zeros(3)" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "SINGLE_LABEL = True" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# simple twitter approach\n", "*for learning emoji usage by single (in the meaning of unconnected) twitter messages*" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## loading train data" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "data_root_folder = \"./data_en/\" # i created a symlink here" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "* get all json files in `data_root_folder`" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ "json_files = sorted(glob.glob(data_root_folder + \"/*.json\"))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "----" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "* so far, only load the first file" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", " | EMOJI | \n", "HASHTAGS | \n", "LINKED_USER | \n", "datetime | \n", "id | \n", "lang | \n", "person | \n", "reply_to | \n", "text | \n", "
---|---|---|---|---|---|---|---|---|---|
0 | \n", "[๐ฅ, ๐] | \n", "[] | \n", "[] | \n", "2017-11-01 13:29:00 | \n", "925716304635547600 | \n", "en | \n", "31507978 | \n", "NaN | \n", "fashionbombdaily's photo <EMOJI><EMOJI>๐ผ | \n", "
1 | \n", "[๐คฆ] | \n", "[] | \n", "[] | \n", "2017-11-01 13:29:00 | \n", "925716304664911900 | \n", "en | \n", "231994649 | \n", "NaN | \n", "Itโs scary how on point my horoscope be <EMOJI... | \n", "
2 | \n", "[๐] | \n", "[] | \n", "[] | \n", "2017-11-01 13:29:03 | \n", "925716317214089200 | \n", "en | \n", "2592765104 | \n", "NaN | \n", "Woooaaaahhh <EMOJI> | \n", "
3 | \n", "[๐ท] | \n", "[] | \n", "[] | \n", "2017-11-01 13:29:04 | \n", "925716321416949800 | \n", "en | \n", "278737933 | \n", "NaN | \n", "<EMOJI> vivalcli: Portraits by Zhao Guojing an... | \n", "
4 | \n", "[๐ฉ, ๐ฉ] | \n", "[] | \n", "[@hiphopphiIes] | \n", "2017-11-01 13:29:06 | \n", "925716329801310200 | \n", "en | \n", "824586253634981900 | \n", "9.257162e+17 | \n", "<USER> i wanna know too<EMOJI><EMOJI> | \n", "
5 | \n", "[๐ญ, ๐] | \n", "[] | \n", "[@WizMommma] | \n", "2017-11-01 13:29:02 | \n", "925716313019965400 | \n", "en | \n", "1581953814 | \n", "9.257088e+17 | \n", "<USER> veda was yoda too <EMOJI><EMOJI> | \n", "
6 | \n", "[๐] | \n", "[] | \n", "[] | \n", "2017-11-01 13:29:05 | \n", "925716325607133200 | \n", "en | \n", "1001999683 | \n", "NaN | \n", "Iโm less stressed about turning 30 now <EMOJI>... | \n", "
7 | \n", "[๐ฏ] | \n", "[] | \n", "[] | \n", "2017-11-01 13:29:07 | \n", "925716334008082400 | \n", "en | \n", "745222369183043600 | \n", "NaN | \n", "Full charged. <EMOJI> | \n", "
8 | \n", "[๐] | \n", "[] | \n", "[@SeaDimon, @lsarsour] | \n", "2017-11-01 13:29:09 | \n", "925716342401052700 | \n", "en | \n", "798557155217539100 | \n", "9.257147e+17 | \n", "<USER> Thatโs part of the problem, (they) <USE... | \n", "
9 | \n", "[๐, ๐ฅ, ๐ข] | \n", "[] | \n", "[@Ian_khetye] | \n", "2017-11-01 13:29:10 | \n", "925716346570240000 | \n", "en | \n", "744396039126421500 | \n", "9.250629e+17 | \n", "<USER> got me emotional there<EMOJI><EMOJI><EM... | \n", "
10 | \n", "[๐ป] | \n", "[] | \n", "[] | \n", "2017-11-01 13:29:13 | \n", "925716359182520300 | \n", "en | \n", "721490010118205400 | \n", "8.965900e+17 | \n", "back to the yellow <EMOJI> | \n", "
11 | \n", "[๐, ๐บ, ๐] | \n", "[] | \n", "[@Dimafadma] | \n", "2017-11-01 13:29:15 | \n", "925716367558545400 | \n", "en | \n", "520536723 | \n", "9.257159e+17 | \n", "<USER> Happy month to you and your loved ones ... | \n", "
12 | \n", "[๐] | \n", "[#mortdale, #partofthefamily, #gorgeousboy] | \n", "[] | \n", "2017-11-01 13:29:16 | \n", "925716371735900200 | \n", "en | \n", "850852815941517300 | \n", "NaN | \n", "Maxx and Patricia. Family hangs at For Good He... | \n", "
13 | \n", "[๐ญ, ๐คฆ] | \n", "[] | \n", "[] | \n", "2017-11-01 13:29:20 | \n", "925716388513230800 | \n", "en | \n", "914145041588867100 | \n", "NaN | \n", "I need to STOP beating myself up with my thoug... | \n", "
14 | \n", "[๐, ๐] | \n", "[#7YearsOfKMH2] | \n", "[] | \n", "2017-11-01 13:29:20 | \n", "925716388525645800 | \n", "en | \n", "2425405622 | \n", "NaN | \n", "Cutest Son <EMOJI>Roll no. 31 <EMOJI> <HASHTAG> | \n", "
15 | \n", "[๐] | \n", "[] | \n", "[] | \n", "2017-11-01 13:29:22 | \n", "925716396931240000 | \n", "en | \n", "4614871873 | \n", "NaN | \n", "By the summer I should have everything up and ... | \n", "
16 | \n", "[๐, ๐ฅ] | \n", "[] | \n", "[] | \n", "2017-11-01 13:29:23 | \n", "925716401125331000 | \n", "en | \n", "2831608345 | \n", "NaN | \n", "I know my English is not that good but that do... | \n", "
17 | \n", "[๐] | \n", "[] | \n", "[@yungbabytate] | \n", "2017-11-01 13:29:23 | \n", "925716401133948900 | \n", "en | \n", "788571974633009200 | \n", "9.255778e+17 | \n", "<USER> I <EMOJI> u mama | \n", "
18 | \n", "[๐] | \n", "[] | \n", "[@cmckenney] | \n", "2017-11-01 13:29:23 | \n", "925716401125544000 | \n", "en | \n", "218307802 | \n", "9.257115e+17 | \n", "<USER> That picture was NOT taken this morning... | \n", "
19 | \n", "[๐ ] | \n", "[#footfetishnation] | \n", "[] | \n", "2017-11-01 13:29:25 | \n", "925716409489002500 | \n", "en | \n", "885261166146179100 | \n", "NaN | \n", "Welcome to <HASHTAG> <EMOJI> | \n", "
20 | \n", "[๐, ๐] | \n", "[] | \n", "[] | \n", "2017-11-01 13:29:25 | \n", "925716409497272300 | \n", "en | \n", "831437760833609700 | \n", "NaN | \n", "Awkward <EMOJI><EMOJI> | \n", "
21 | \n", "[๐ค, ๐บ] | \n", "[] | \n", "[] | \n", "2017-11-01 13:29:26 | \n", "925716413699854300 | \n", "en | \n", "231664542 | \n", "NaN | \n", "back at it with supernatural <EMOJI><EMOJI> | \n", "
22 | \n", "[๐ฏ] | \n", "[] | \n", "[] | \n", "2017-11-01 13:29:26 | \n", "925716413679009800 | \n", "en | \n", "3196847035 | \n", "NaN | \n", "One of the best things I've learned was to sto... | \n", "
23 | \n", "[๐ , ๐ฆ, ๐] | \n", "[] | \n", "[@ctrlpurp] | \n", "2017-11-01 13:29:29 | \n", "925716426278735900 | \n", "en | \n", "918492858352635900 | \n", "9.257161e+17 | \n", "<USER> Can I taste?<EMOJI><EMOJI><EMOJI> | \n", "
24 | \n", "[๐] | \n", "[] | \n", "[@saunders_court1] | \n", "2017-11-01 13:29:30 | \n", "925716430473039900 | \n", "en | \n", "3471187337 | \n", "9.257163e+17 | \n", "<USER> we miss you โน๏ธ<EMOJI> | \n", "
25 | \n", "[๐ค, ๐ค, ๐ค] | \n", "[] | \n", "[] | \n", "2017-11-01 13:29:31 | \n", "925716434667184100 | \n", "en | \n", "780060488600199200 | \n", "NaN | \n", "Actually my bias in WJSN are Eunseo & Bona... | \n", "
26 | \n", "[๐ด] | \n", "[] | \n", "[] | \n", "2017-11-01 13:29:32 | \n", "925716438853345300 | \n", "en | \n", "388380690 | \n", "NaN | \n", "I so cannot be bothered with the rest of the d... | \n", "
27 | \n", "[๐] | \n", "[] | \n", "[@xxxtentacion] | \n", "2017-11-01 13:29:35 | \n", "925716451457163300 | \n", "en | \n", "899320696869974000 | \n", "NaN | \n", "<USER> 2lit4life<EMOJI> | \n", "
28 | \n", "[๐, ๐] | \n", "[] | \n", "[] | \n", "2017-11-01 13:29:35 | \n", "925716451461357600 | \n", "en | \n", "784790670 | \n", "NaN | \n", "Iโm not stop saying that!<EMOJI><EMOJI> | \n", "
29 | \n", "[๐, ๐, ๐, ๐, ๐, ๐] | \n", "[] | \n", "[@justinerooney_] | \n", "2017-11-01 13:29:37 | \n", "925716459828936700 | \n", "en | \n", "3051266655 | \n", "NaN | \n", "<USER> HAPPY BIHDAY <EMOJI><EMOJI><EMOJI><EMOJ... | \n", "
... | \n", "... | \n", "... | \n", "... | \n", "... | \n", "... | \n", "... | \n", "... | \n", "... | \n", "... | \n", "
68703 | \n", "[๐] | \n", "[#halloweencostumes] | \n", "[] | \n", "2017-11-01 07:23:04 | \n", "925624214522036200 | \n", "en | \n", "1672876458 | \n", "NaN | \n", "<HASHTAG> this one falls under the weird crazy... | \n", "
68704 | \n", "[๐, ๐ฉ] | \n", "[] | \n", "[] | \n", "2017-11-01 07:23:05 | \n", "925624218682777600 | \n", "en | \n", "382473866 | \n", "NaN | \n", "I'm not allowed to have chocolates yet, then I... | \n", "
68705 | \n", "[๐] | \n", "[#MUFC] | \n", "[] | \n", "2017-11-01 07:23:06 | \n", "925624222889766900 | \n", "en | \n", "893145405457911800 | \n", "NaN | \n", "Manchester United manager Mourinho slams 'spec... | \n", "
68706 | \n", "[๐] | \n", "[] | \n", "[] | \n", "2017-11-01 07:23:07 | \n", "925624227088121900 | \n", "en | \n", "240378516 | \n", "NaN | \n", "<EMOJI> en Bushwhick | \n", "
68707 | \n", "[๐, ๐, ๐, ๐] | \n", "[#BurkeCentre] | \n", "[] | \n", "2017-11-01 07:23:18 | \n", "925624273237983200 | \n", "en | \n", "784620573209002000 | \n", "NaN | \n", ": <HASHTAG> <EMOJI> <EMOJI><EMOJI> For Taxi <E... | \n", "
68708 | \n", "[๐] | \n", "[] | \n", "[@mychosliaheart, @BarrettoJulia, @iamjoshuaga... | \n", "2017-11-01 07:23:18 | \n", "925624273212805100 | \n", "en | \n", "170998187 | \n", "9.254136e+17 | \n", "<USER> <USER> <USER> Look, Mammeh and Daddeh! ... | \n", "
68709 | \n", "[๐] | \n", "[] | \n", "[] | \n", "2017-11-01 07:23:24 | \n", "925624298395533300 | \n", "en | \n", "1348667816 | \n", "NaN | \n", "Life is so good with you <EMOJI> | \n", "
68710 | \n", "[๐, ๐, ๐] | \n", "[#portlandoregon, #portlandhalloween, #carrieโฆ] | \n", "[] | \n", "2017-11-01 07:23:24 | \n", "925624298378801200 | \n", "en | \n", "722481645765300200 | \n", "NaN | \n", "Happy Halloween! <EMOJI>๐ฝ<EMOJI><EMOJI> <HASHT... | \n", "
68711 | \n", "[๐] | \n", "[] | \n", "[@8limbsbondi...] | \n", "2017-11-01 07:23:26 | \n", "925624306779897900 | \n", "en | \n", "2443251500 | \n", "NaN | \n", "Some work on the ropes in todayโs boxing class... | \n", "
68712 | \n", "[๐ญ] | \n", "[] | \n", "[] | \n", "2017-11-01 07:23:27 | \n", "925624310974136300 | \n", "en | \n", "2406186390 | \n", "NaN | \n", "Scotty and Kristenโs halloween costumes <EMOJI> | \n", "
68713 | \n", "[๐] | \n", "[] | \n", "[@rfrandrea, @AdaaanAndyyy] | \n", "2017-11-01 07:23:30 | \n", "925624323557146600 | \n", "en | \n", "1039448149 | \n", "9.256150e+17 | \n", "<USER> <USER> May pre-month celebration sis <E... | \n", "
68714 | \n", "[๐ญ, ๐ญ, ๐] | \n", "[] | \n", "[@peachshua1230] | \n", "2017-11-01 07:23:31 | \n", "925624327755591700 | \n", "en | \n", "845085544589672400 | \n", "9.256009e+17 | \n", "<USER> Awww <EMOJI> Ajsksjdjd im smiling like ... | \n", "
68715 | \n", "[๐ซ, โ] | \n", "[] | \n", "[] | \n", "2017-11-01 07:23:34 | \n", "925624340342812700 | \n", "en | \n", "924752524871131100 | \n", "NaN | \n", "I hate when I send a text or snap n I'm so anx... | \n", "
68716 | \n", "[๐, ๐] | \n", "[] | \n", "[] | \n", "2017-11-01 07:23:34 | \n", "925624340355280900 | \n", "en | \n", "419493819 | \n", "NaN | \n", "The answer is no I have no plans and I never l... | \n", "
68717 | \n", "[๐ญ] | \n", "[] | \n", "[@BeachBoy_Gab] | \n", "2017-11-01 07:23:34 | \n", "925624340346937300 | \n", "en | \n", "2457745952 | \n", "9.256219e+17 | \n", "<USER> LMAOOO I'm so proud <EMOJI> | \n", "
68718 | \n", "[๐ข, ๐] | \n", "[] | \n", "[] | \n", "2017-11-01 07:23:34 | \n", "925624340338507800 | \n", "en | \n", "1955767531 | \n", "NaN | \n", "My cousin/little sister is leaving to San Fran... | \n", "
68719 | \n", "[๐] | \n", "[] | \n", "[] | \n", "2017-11-01 07:23:35 | \n", "925624344524361700 | \n", "en | \n", "796490344581898200 | \n", "NaN | \n", "Can't be alone w my thoughts tonight so just g... | \n", "
68720 | \n", "[๐, ๐, ๐, ๐, ๐] | \n", "[] | \n", "[] | \n", "2017-11-01 07:23:36 | \n", "925624348710285300 | \n", "en | \n", "907808317124177900 | \n", "NaN | \n", "<EMOJI><EMOJI><EMOJI><EMOJI><EMOJI> ambot!!! | \n", "
68721 | \n", "[๐, ๐, ๐] | \n", "[] | \n", "[] | \n", "2017-11-01 07:23:37 | \n", "925624352929910800 | \n", "en | \n", "262162415 | \n", "NaN | \n", "<EMOJI>๐ฝ<EMOJI>๐ฝ 1st of the month!!Happy 1st o... | \n", "
68722 | \n", "[๐ด] | \n", "[#WednesdayWisdom] | \n", "[] | \n", "2017-11-01 07:23:41 | \n", "925624369715515400 | \n", "en | \n", "574882525 | \n", "NaN | \n", "<HASHTAG> ... stay in bed <EMOJI> | \n", "
68723 | \n", "[๐, ๐, ๐, ๐] | \n", "[] | \n", "[@Louis_Tomlinson, @NiallOfficial] | \n", "2017-11-01 07:23:44 | \n", "925624382269124600 | \n", "en | \n", "556175173 | \n", "9.254038e+17 | \n", "<USER> <USER> THIS IS GOLD. GOLD. <EMOJI><EMOJ... | \n", "
68724 | \n", "[๐] | \n", "[] | \n", "[] | \n", "2017-11-01 07:23:45 | \n", "925624386455031800 | \n", "en | \n", "1610265588 | \n", "NaN | \n", "<EMOJI> thank you for the kind compliment | \n", "
68725 | \n", "[๐] | \n", "[] | \n", "[] | \n", "2017-11-01 07:23:45 | \n", "925624386454937600 | \n", "en | \n", "4760724450 | \n", "NaN | \n", "Enjoyed the silence <EMOJI> | \n", "
68726 | \n", "[โจ] | \n", "[] | \n", "[] | \n", "2017-11-01 07:23:46 | \n", "925624390657572900 | \n", "en | \n", "882858115636514800 | \n", "NaN | \n", "OS: Spiderman Homecoming <EMOJI> | \n", "
68727 | \n", "[๐คท] | \n", "[] | \n", "[] | \n", "2017-11-01 07:23:46 | \n", "925624390682849300 | \n", "en | \n", "188129628 | \n", "NaN | \n", "Go to hell <EMOJI>๐ฝโโ๏ธ | \n", "
68728 | \n", "[๐] | \n", "[] | \n", "[] | \n", "2017-11-01 07:23:46 | \n", "925624390666129400 | \n", "en | \n", "2473135939 | \n", "NaN | \n", "Thank you Yomi! <EMOJI> | \n", "
68729 | \n", "[๐] | \n", "[] | \n", "[@discopiggu] | \n", "2017-11-01 07:23:46 | \n", "925624390670106600 | \n", "en | \n", "2373584209 | \n", "9.256241e+17 | \n", "<USER> Lol. Just enjoy the stars. Music Kidhar... | \n", "
68730 | \n", "[๐] | \n", "[#NYCStrong] | \n", "[] | \n", "2017-11-01 07:23:50 | \n", "925624407459971100 | \n", "en | \n", "181689756 | \n", "NaN | \n", "Thoughts and prayers for NY<EMOJI>๐ป <HASHTAG> | \n", "
68731 | \n", "[๐] | \n", "[#GreatSuccess] | \n", "[@BrianyH] | \n", "2017-11-01 07:23:50 | \n", "925624407460057100 | \n", "en | \n", "60160788 | \n", "9.254610e+17 | \n", "<USER> I searched COCK, PENIS, SHLONG, WINKY, ... | \n", "
68732 | \n", "[๐, ๐ป, ๐ป, ๐, ๐, ๐, ๐, ๐, ๐] | \n", "[] | \n", "[@amitbarman520] | \n", "2017-11-01 07:23:53 | \n", "925624420022063100 | \n", "en | \n", "3792290725 | \n", "9.256215e+17 | \n", "<USER> Thank you so much<EMOJI><EMOJI><EMOJI><... | \n", "
68733 rows ร 9 columns
\n", "\n", " | predict | \n", "teacher | \n", "text | \n", "
---|---|---|---|
15278 | \n", "(๐ ,) | \n", "(๐ ,) | \n", "<USER> <USER> 2 years? โ | \n", "
36882 | \n", "(๐,) | \n", "(๐,) | \n", "tell me why this is me everyday | \n", "
65543 | \n", "(๐,) | \n", "(๐,) | \n", "please follow us ::Twitter โถ๏ธ โถ๏ธ โถ๏ธMARSHMALL... | \n", "
65893 | \n", "(๐,) | \n", "(๐,) | \n", "This | \n", "
55496 | \n", "(๐ณ,) | \n", "(๐,) | \n", "<USER> Im in love | \n", "
61997 | \n", "(๐,) | \n", "(๐ธ,) | \n", "<USER> Miss you gemmthankyou sooo muchโคโค | \n", "
15819 | \n", "(๐น,) | \n", "(๐,) | \n", "Me as a parent | \n", "
21340 | \n", "(๐,) | \n", "(๐,) | \n", "Really betting my manager rn for tonightโs game | \n", "
26691 | \n", "(๐ณ,) | \n", "(๐ฉ,) | \n", "Y ES IM A BAD BOY SO I LIKE BAD GIRL | \n", "
57025 | \n", "(๐ท,) | \n", "(๐,) | \n", "<USER> Just downloaded to my fire stick over ... | \n", "
17967 | \n", "(๐,) | \n", "(๐,) | \n", "<USER> Love this ! Hollywood is the man | \n", "
68555 | \n", "(๐,) | \n", "(๐,) | \n", "<USER> Absolutely but if it had been a banana ... | \n", "
20784 | \n", "(๐,) | \n", "(๐ป,) | \n", "La Grasaaaaaaaaa Reverse Shatter Back Board S... | \n", "
68300 | \n", "(๐ฅฃ,) | \n", "(๐,) | \n", "I go to New York next month ahhhhhh | \n", "
4659 | \n", "(๐,) | \n", "(๐ญ,) | \n", "<USER> Iโm trying to enjoy it as much as the f... | \n", "
66706 | \n", "(๐,) | \n", "(๐,) | \n", ": Salamander Resort <HASHTAG> ๐ฝ๏ธ For Taxi 70... | \n", "
47510 | \n", "(๐,) | \n", "(๐,) | \n", "For your actions there WILL be consequences. Y... | \n", "
55294 | \n", "(๐พ,) | \n", "(๐,) | \n", "<USER> Omg my cousin charging 40 | \n", "
621 | \n", "(๐,) | \n", "(๐คฆ,) | \n", "Zoey daycare lets her do whatever she want, I ... | \n", "
31011 | \n", "(๐,) | \n", "(๐,) | \n", "yeah! โ๏ธ | \n", "
30659 | \n", "(๐,) | \n", "(๐ช,) | \n", "This was so much I mean SO MUCH over due for t... | \n", "
66284 | \n", "(๐ค,) | \n", "(๐,) | \n", "Parent contact booking is now open! Please fol... | \n", "
66758 | \n", "(๐,) | \n", "(๐ฅ,) | \n", "<USER> OCONUS we donโt have the Turkey Day buf... | \n", "
13456 | \n", "(๐ ,) | \n", "(๐,) | \n", "Dems realtor business partners have lots of ni... | \n", "
15057 | \n", "(๐ฅฃ,) | \n", "(๐,) | \n", "Ready for the dodgers fans excuses | \n", "
188 | \n", "(๐,) | \n", "(๐,) | \n", "<USER> Happy Birthday Angge | \n", "
5054 | \n", "(๐,) | \n", "(๐,) | \n", "<USER> <USER> Lawsuit | \n", "
18099 | \n", "(๐,) | \n", "(๐ค,) | \n", "Finna roll this up no excuses | \n", "
17120 | \n", "(๐,) | \n", "(โก,) | \n", "But itโs forever fuck em and keep it gutta ๏ธ | \n", "
14552 | \n", "(๐,) | \n", "(๐,) | \n", "I just need to hubby a hockey player up | \n", "
... | \n", "... | \n", "... | \n", "... | \n", "
21819 | \n", "(๐,) | \n", "(๐,) | \n", "Niggas funny af | \n", "
18460 | \n", "(โฟ,) | \n", "(๐,) | \n", "1. Donโt EVER disrespect Thanksgiving cuz yo f... | \n", "
60291 | \n", "(๐ฎ,) | \n", "(๐,) | \n", "happy bday lexi <USER> | \n", "
22623 | \n", "(๐,) | \n", "(๐ฉ,) | \n", "Mr.Lazaro said โthereโs my girlsโ when he seen... | \n", "
87 | \n", "(๐,) | \n", "(๐,) | \n", "Happy Birthday Blake!!!! I hope you have an am... | \n", "
2884 | \n", "(๐,) | \n", "(๐,) | \n", "<USER> This would be a good time to do a Week ... | \n", "
30212 | \n", "(๐ง,) | \n", "(๐,) | \n", "<USER> Done, pls โค๏ธ y FAV | \n", "
55813 | \n", "(๐พ,) | \n", "(๐,) | \n", "There is just something about <USER> <HASHTAG> | \n", "
11657 | \n", "(๐,) | \n", "(๐ด,) | \n", "<USER> you're brudda! | \n", "
18215 | \n", "(๐น,) | \n", "(๐คท,) | \n", "Not being depressed โ๏ธ | \n", "
48609 | \n", "(๐ง,) | \n", "(๐ค,) | \n", "<USER> I โค you. My woman crush everyday as alw... | \n", "
16079 | \n", "(๐ฅฃ,) | \n", "(โ,) | \n", "I added a video to a <USER> playlist ๐ณ ๏ธ The ... | \n", "
28380 | \n", "(๐ฅฃ,) | \n", "(๐,) | \n", "Dodgers | \n", "
41212 | \n", "(๐,) | \n", "(๐ด,) | \n", "[LIVE] on <USER> | \n", "
67555 | \n", "(๐,) | \n", "(๐,) | \n", "โข <USER> my precious Niya always checking up o... | \n", "
45490 | \n", "(๐,) | \n", "(๐คท,) | \n", "If you wanna make things right it go both ways โ๏ธ | \n", "
54052 | \n", "(๐,) | \n", "(๐ข,) | \n", "Poor baby!!! | \n", "
54517 | \n", "(๐น,) | \n", "(๐,) | \n", "<USER> happy birthday, hope you have a good day | \n", "
24047 | \n", "(๐ท,) | \n", "(๐,) | \n", "Donโt get mad when them haters stare let em lo... | \n", "
253 | \n", "(๐ฅ,) | \n", "(๐,) | \n", "Sag season right around the corner. 25 on the ... | \n", "
64707 | \n", "(๐ธ,) | \n", "(๐,) | \n", "<USER> Letโs see his costume | \n", "
34279 | \n", "(๐ฏ,) | \n", "(๐ฅ,) | \n", "<USER> I AM NERVOUS!! | \n", "
58342 | \n", "(๐ค,) | \n", "(๐,) | \n", "<USER> This looks amazing | \n", "
16361 | \n", "(๐,) | \n", "(๐,) | \n", "<USER> <USER> <USER> <USER> <USER> <USER> Shit... | \n", "
39069 | \n", "(๐,) | \n", "(๐,) | \n", "THIS | \n", "
59739 | \n", "(๐ท,) | \n", "(๐,) | \n", "I've never consumed a reindeer. | \n", "
49140 | \n", "(๐,) | \n", "(๐ค,) | \n", "Sales is the heart of business. | \n", "
47901 | \n", "(๐,) | \n", "(๐ญ,) | \n", "bruuuuhhh | \n", "
4241 | \n", "(๐ฒ,) | \n", "(๐,) | \n", "<USER> Congratulations ! youโve graduated with... | \n", "
24123 | \n", "(โ,) | \n", "(๐,) | \n", "<USER> Your A Rock Star! How You Put Up With ... | \n", "
6874 rows ร 3 columns
\n", "Failed to display Jupyter Widget of type Text
.
\n", " If you're reading this message in the Jupyter Notebook or JupyterLab Notebook, it may mean\n", " that the widgets JavaScript is still loading. If this message persists, it\n", " likely means that the widgets JavaScript library is either not installed or\n", " not enabled. See the Jupyter\n", " Widgets Documentation for setup instructions.\n", "
\n", "\n", " If you're reading this message in another frontend (for example, a static\n", " rendering on GitHub or NBViewer),\n", " it may mean that your frontend doesn't currently support widgets.\n", "
\n" ], "text/plain": [ "Text(value='')" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "f3af634eb2e64822921b57848e032dcf", "version_major": 2, "version_minor": 0 }, "text/html": [ "Failed to display Jupyter Widget of type VBox
.
\n", " If you're reading this message in the Jupyter Notebook or JupyterLab Notebook, it may mean\n", " that the widgets JavaScript is still loading. If this message persists, it\n", " likely means that the widgets JavaScript library is either not installed or\n", " not enabled. See the Jupyter\n", " Widgets Documentation for setup instructions.\n", "
\n", "\n", " If you're reading this message in another frontend (for example, a static\n", " rendering on GitHub or NBViewer),\n", " it may mean that your frontend doesn't currently support widgets.\n", "
\n" ], "text/plain": [ "VBox(children=(Button(description='get smiley', icon='check', style=ButtonStyle(), tooltip='Click me'), Output()))" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "out = widgets.Output()\n", "\n", "t = widgets.Text()\n", "b = widgets.Button(\n", " description='get smiley',\n", " disabled=False,\n", " button_style='', # 'success', 'info', 'warning', 'danger' or ''\n", " tooltip='Click me',\n", " icon='check'\n", ")\n", "\n", "\n", "\n", "def handle_submit(sender):\n", " with out:\n", " clear_output()\n", " with out:\n", " pred = clf.predict(vectorizer.transform([t.value]))\n", " \n", " #binary_pred = np.zeros_like(pred)\n", " #binary_pred[np.arange(len(pred)), pred.argmax(1)] = 1\n", " \n", " binary_pred = []\n", " \n", " s = np.argsort(pred[0])[-5:]\n", " \n", " for i in s:\n", " b = np.zeros_like(pred[0])\n", " b[i] = 1\n", " binary_pred.append(b)\n", " \n", " binary_pred = np.array(binary_pred)\n", " \n", " display(Markdown(\"# \" + str((mlb_single if SINGLE_LABEL else mlb).inverse_transform(binary_pred))))\n", "\n", "b.on_click(handle_submit)\n", " \n", "display(t)\n", "display(widgets.VBox([b, out])) " ] }, { "cell_type": "code", "execution_count": 44, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "0.9991493705341953\n", "15\n" ] } ], "source": [ "import numpy as np\n", "\n", "y_trans = mlb.inverse_transform(yt1)\n", "pred_trans = mlb.inverse_transform(yt1)\n", "\n", "# evaluate accuracy\n", "pos = 0\n", "neg = 0\n", "all = 0\n", "for entry in range(len(y_trans)):\n", " if len(np.intersect1d(y_trans[entry], pred_trans[entry])) > 0:\n", " pos += 1\n", " else:\n", " neg += 1\n", " all += 1\n", "print(pos/all)\n", "print(neg)" ] }, { "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": "3.6.5" } }, "nbformat": 4, "nbformat_minor": 2 }