Spaces:
Running
Running
File size: 87,013 Bytes
03287bc |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 |
{
"cells": [
{
"cell_type": "markdown",
"id": "8ea54fcd-ef4a-42cb-ae26-cbdc6f6ffc64",
"metadata": {
"tags": []
},
"source": [
"# Duct Tape Pipeline\n",
"To explore how users may interact with interactive visualizations of counterfactuals for evolving the Interactive Model Card, we will need to first find a way to generate counterfactuals based on a given input. We want the user to be able to provide their input and direct the system to generate counterfactuals based on a part of speech that is significant to the model. The system should then provide a data frame of counterfactuals to be used in an interactive visualization. Below is an example wireframe of the experience based on previous research.\n",
"\n",
"\n",
"\n",
"## Goals of this notebook\n",
"* Clean up the flow in the \"duct tape pipeline\".\n",
"* See if I can extract the LIME list for visualization"
]
},
{
"cell_type": "markdown",
"id": "736e6375-dd6d-4188-b8b1-92bded2bcd02",
"metadata": {},
"source": [
"## Loading the libraries and models"
]
},
{
"cell_type": "code",
"execution_count": 1,
"id": "7f581785-e642-4f74-9f67-06a63820eaf2",
"metadata": {},
"outputs": [],
"source": [
"#Import the libraries we know we'll need for the Generator.\n",
"import pandas as pd, spacy, nltk, numpy as np\n",
"from spacy import displacy\n",
"from spacy.matcher import Matcher\n",
"#!python -m spacy download en_core_web_sm\n",
"nlp = spacy.load(\"en_core_web_md\")\n",
"lemmatizer = nlp.get_pipe(\"lemmatizer\")\n",
"\n",
"#Import the libraries to support the model, predictions, and LIME.\n",
"from transformers import AutoTokenizer, AutoModelForSequenceClassification, TextClassificationPipeline\n",
"import lime\n",
"import torch\n",
"import torch.nn.functional as F\n",
"from lime.lime_text import LimeTextExplainer\n",
"\n",
"#Import the libraries for generating interactive visualizations.\n",
"import altair as alt"
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "cbe2b292-e33e-4915-8e61-bba5327fb643",
"metadata": {},
"outputs": [],
"source": [
"#Defining all necessary variables and instances.\n",
"tokenizer = AutoTokenizer.from_pretrained(\"distilbert-base-uncased-finetuned-sst-2-english\")\n",
"model = AutoModelForSequenceClassification.from_pretrained(\"distilbert-base-uncased-finetuned-sst-2-english\")\n",
"class_names = ['negative', 'positive']\n",
"explainer = LimeTextExplainer(class_names=class_names)"
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "197c3e26-0fdf-49c6-9135-57f1fd55d3e3",
"metadata": {},
"outputs": [],
"source": [
"#Defining a Predictor required for LIME to function.\n",
"def predictor(texts):\n",
" outputs = model(**tokenizer(texts, return_tensors=\"pt\", padding=True))\n",
" probas = F.softmax(outputs.logits, dim=1).detach().numpy()\n",
" return probas"
]
},
{
"cell_type": "code",
"execution_count": 4,
"id": "013af6ac-f7d1-41d2-a601-b0f9a4870815",
"metadata": {},
"outputs": [],
"source": [
"#Instantiate a matcher and use it to test some patterns.\n",
"matcher = Matcher(nlp.vocab)\n",
"pattern = [{\"ENT_TYPE\": {\"IN\":[\"NORP\",\"GPE\"]}}]\n",
"matcher.add(\"proper_noun\", [pattern])\n",
"pattern_test = [{\"DEP\": \"amod\"},{\"DEP\":\"attr\"},{\"TEXT\":\"-\"},{\"DEP\":\"attr\",\"OP\":\"+\"}]\n",
"matcher.add(\"amod_attr\",[pattern_test])\n",
"pattern_an = [{\"DEP\": \"amod\"},{\"POS\":{\"IN\":[\"NOUN\",\"PROPN\"]}},{\"DEP\":{\"NOT_IN\":[\"attr\"]}}]\n",
"matcher.add(\"amod_noun\", [pattern_an])"
]
},
{
"cell_type": "code",
"execution_count": 5,
"id": "f6ac821d-7b56-446e-b9ca-42a5f5afd198",
"metadata": {},
"outputs": [],
"source": [
"def match_this(matcher, doc):\n",
" matches = matcher(doc)\n",
" for match_id, start, end in matches:\n",
" matched_span = doc[start:end]\n",
" print(f\"Mached {matched_span.text} by the rule {nlp.vocab.strings[match_id]}.\")\n",
" return matches"
]
},
{
"cell_type": "markdown",
"id": "c23d48c4-f5ab-4428-9244-0786e9903a8e",
"metadata": {
"tags": []
},
"source": [
"## Building the Duct-Tape Pipeline cell-by-cell"
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "a373fc00-401a-4def-9f09-de73d485ac13",
"metadata": {},
"outputs": [],
"source": [
"gender = [\"man\", \"woman\",\"girl\",\"boy\",\"male\",\"female\",\"husband\",\"wife\",\"girlfriend\",\"boyfriend\",\"brother\",\"sister\",\"aunt\",\"uncle\",\"grandma\",\"grandpa\",\"granny\",\"granps\",\"grandmother\",\"grandfather\",\"mama\",\"dada\",\"Ma\",\"Pa\",\"lady\",\"gentleman\"]"
]
},
{
"cell_type": "code",
"execution_count": 7,
"id": "8b02a5d4-8a6b-4e5e-8f15-4f9182fe341f",
"metadata": {},
"outputs": [],
"source": [
"def select_crit(document, options=False, limelist=False):\n",
" '''This function is meant to select the critical part of a sentence. Critical, in this context means\n",
" the part of the sentence that is either: A) a PROPN from the correct entity group; B) an ADJ associated with a NOUN;\n",
" C) a NOUN that represents gender. It also checks this against what the model thinks is important if the user defines \"options\" as \"LIME\" or True.'''\n",
" chunks = list(document.noun_chunks)\n",
" pos_options = []\n",
" lime_options = []\n",
" \n",
" #Identify what the model cares about.\n",
" if options:\n",
" exp = explainer.explain_instance(document.text, predictor, num_features=15, num_samples=2000)\n",
" lime_results = exp.as_list()\n",
" #prints the results from lime for QA.\n",
" if limelist == True:\n",
" print(lime_results)\n",
" for feature in lime_results:\n",
" lime_options.append(feature[0])\n",
" lime_results = pd.DataFrame(lime_results, columns=[\"Word\",\"Weight\"])\n",
" \n",
" #Identify what we care about \"parts of speech\"\n",
" for chunk in chunks:\n",
" #The use of chunk[-1] is due to testing that it appears to always match the root\n",
" root = chunk[-1]\n",
" #This currently matches to a list I've created. I don't know the best way to deal with this so I'm leaving it as is for the moment.\n",
" if root.text.lower() in gender:\n",
" cur_values = [token.text for token in chunk if token.pos_ in [\"NOUN\",\"ADJ\"]]\n",
" if (all(elem in lime_options for elem in cur_values) and ((options == \"LIME\") or (options == True))) or ((options != \"LIME\") and (options != True)):\n",
" pos_options.extend(cur_values)\n",
" #print(f\"From {chunk.text}, {cur_values} added to pos_options due to gender.\") #for QA\n",
" #This is currently set to pick up entities in a particular set of groups (which I recently expanded). Should it just pick up all named entities?\n",
" elif root.ent_type_ in [\"GPE\",\"NORP\",\"DATE\",\"EVENT\"]:\n",
" cur_values = []\n",
" if (len(chunk) > 1) and (chunk[-2].dep_ == \"compound\"):\n",
" #creates the compound element of the noun\n",
" compound = [x.text for x in chunk if x.dep_ == \"compound\"]\n",
" print(f\"This is the contents of {compound} and it is {all(elem in lime_options for elem in compound)} that all elements are present in {lime_options}.\") #for QA\n",
" #checks to see all elements in the compound are important to the model or use the compound if not checking importance.\n",
" if (all(elem in lime_options for elem in compound) and ((options == \"LIME\") or (options == True))) or ((options != \"LIME\") and (options != True)):\n",
" #creates a span for the entirety of the compound noun and adds it to the list.\n",
" span = -1 * (1 + len(compound))\n",
" pos_options.append(chunk[span:].text)\n",
" cur_values + [token.text for token in chunk if token.pos_ == \"ADJ\"]\n",
" else: \n",
" cur_values = [token.text for token in chunk if (token.ent_type_ in [\"GPE\",\"NORP\",\"DATE\",\"EVENT\"]) or (token.pos_ == \"ADJ\")]\n",
" if (all(elem in lime_options for elem in cur_values) and ((options == \"LIME\") or (options == True))) or ((options != \"LIME\") and (options != True)):\n",
" pos_options.extend(cur_values)\n",
" print(f\"From {chunk.text}, {cur_values} and {pos_options} added to pos_options due to entity recognition.\") #for QA\n",
" elif len(chunk) > 1:\n",
" cur_values = [token.text for token in chunk if token.pos_ in [\"NOUN\",\"ADJ\"]]\n",
" if (all(elem in lime_options for elem in cur_values) and ((options == \"LIME\") or (options == True))) or ((options != \"LIME\") and (options != True)):\n",
" pos_options.extend(cur_values)\n",
" print(f\"From {chunk.text}, {cur_values} added to pos_options due to wildcard.\") #for QA\n",
" else:\n",
" print(f\"No options added for \\'{chunk.text}\\' \")\n",
" \n",
" \n",
" #Return the correct set of options based on user input, defaults to POS for simplicity.\n",
" if options == \"LIME\":\n",
" return pos_options, lime_results\n",
" else:\n",
" return pos_options"
]
},
{
"cell_type": "code",
"execution_count": 8,
"id": "d43e202e-64b9-4cea-b117-82492c9ee5f4",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"From This film, ['film'] added to pos_options due to wildcard.\n",
"From Iraq, ['Iraq'] and ['film', 'Iraq'] added to pos_options due to entity recognition.\n"
]
}
],
"source": [
"#Test to make sure all three options work\n",
"text4 = \"This film was filmed in Iraq.\"\n",
"doc4 = nlp(text4)\n",
"lime4, limedf = select_crit(doc4,options=\"LIME\")"
]
},
{
"cell_type": "code",
"execution_count": 9,
"id": "a0e55a24-65df-429e-a0cd-8daf91a5d242",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"<div id=\"altair-viz-23e37c16acf34cbead4ebdbe2bddfdb5\"></div>\n",
"<script type=\"text/javascript\">\n",
" var VEGA_DEBUG = (typeof VEGA_DEBUG == \"undefined\") ? {} : VEGA_DEBUG;\n",
" (function(spec, embedOpt){\n",
" let outputDiv = document.currentScript.previousElementSibling;\n",
" if (outputDiv.id !== \"altair-viz-23e37c16acf34cbead4ebdbe2bddfdb5\") {\n",
" outputDiv = document.getElementById(\"altair-viz-23e37c16acf34cbead4ebdbe2bddfdb5\");\n",
" }\n",
" const paths = {\n",
" \"vega\": \"https://cdn.jsdelivr.net/npm//vega@5?noext\",\n",
" \"vega-lib\": \"https://cdn.jsdelivr.net/npm//vega-lib?noext\",\n",
" \"vega-lite\": \"https://cdn.jsdelivr.net/npm//vega-lite@4.17.0?noext\",\n",
" \"vega-embed\": \"https://cdn.jsdelivr.net/npm//vega-embed@6?noext\",\n",
" };\n",
"\n",
" function maybeLoadScript(lib, version) {\n",
" var key = `${lib.replace(\"-\", \"\")}_version`;\n",
" return (VEGA_DEBUG[key] == version) ?\n",
" Promise.resolve(paths[lib]) :\n",
" new Promise(function(resolve, reject) {\n",
" var s = document.createElement('script');\n",
" document.getElementsByTagName(\"head\")[0].appendChild(s);\n",
" s.async = true;\n",
" s.onload = () => {\n",
" VEGA_DEBUG[key] = version;\n",
" return resolve(paths[lib]);\n",
" };\n",
" s.onerror = () => reject(`Error loading script: ${paths[lib]}`);\n",
" s.src = paths[lib];\n",
" });\n",
" }\n",
"\n",
" function showError(err) {\n",
" outputDiv.innerHTML = `<div class=\"error\" style=\"color:red;\">${err}</div>`;\n",
" throw err;\n",
" }\n",
"\n",
" function displayChart(vegaEmbed) {\n",
" vegaEmbed(outputDiv, spec, embedOpt)\n",
" .catch(err => showError(`Javascript Error: ${err.message}<br>This usually means there's a typo in your chart specification. See the javascript console for the full traceback.`));\n",
" }\n",
"\n",
" if(typeof define === \"function\" && define.amd) {\n",
" requirejs.config({paths});\n",
" require([\"vega-embed\"], displayChart, err => showError(`Error loading script: ${err.message}`));\n",
" } else {\n",
" maybeLoadScript(\"vega\", \"5\")\n",
" .then(() => maybeLoadScript(\"vega-lite\", \"4.17.0\"))\n",
" .then(() => maybeLoadScript(\"vega-embed\", \"6\"))\n",
" .catch(showError)\n",
" .then(() => displayChart(vegaEmbed));\n",
" }\n",
" })({\"config\": {\"view\": {\"continuousWidth\": 400, \"continuousHeight\": 300, \"strokeWidth\": 0}, \"axis\": {\"grid\": false}}, \"layer\": [{\"mark\": \"bar\", \"encoding\": {\"color\": {\"field\": \"Weight\", \"legend\": null, \"scale\": {\"domain\": [0], \"range\": \"diverging\", \"scheme\": \"blueorange\", \"type\": \"threshold\"}, \"type\": \"quantitative\"}, \"tooltip\": [{\"field\": \"Word\", \"type\": \"nominal\"}, {\"field\": \"Weight\", \"type\": \"quantitative\"}], \"x\": {\"field\": \"Weight\", \"scale\": {\"domain\": [-1, 1]}, \"type\": \"quantitative\"}, \"y\": {\"axis\": null, \"field\": \"Word\", \"sort\": \"x\", \"type\": \"nominal\"}}, \"title\": \"Importance of individual words\"}, {\"mark\": {\"type\": \"text\", \"align\": \"right\", \"baseline\": \"middle\", \"fill\": \"black\"}, \"encoding\": {\"color\": {\"field\": \"Weight\", \"legend\": null, \"scale\": {\"domain\": [0], \"range\": \"diverging\", \"scheme\": \"blueorange\", \"type\": \"threshold\"}, \"type\": \"quantitative\"}, \"text\": {\"field\": \"Word\", \"type\": \"nominal\"}, \"tooltip\": [{\"field\": \"Word\", \"type\": \"nominal\"}, {\"field\": \"Weight\", \"type\": \"quantitative\"}], \"x\": {\"field\": \"Weight\", \"scale\": {\"domain\": [-1, 1]}, \"type\": \"quantitative\"}, \"y\": {\"axis\": null, \"field\": \"Word\", \"sort\": \"x\", \"type\": \"nominal\"}}, \"title\": \"Importance of individual words\"}], \"data\": {\"name\": \"data-1b001587c028498e70538ed310063e51\"}, \"width\": 300, \"$schema\": \"https://vega.github.io/schema/vega-lite/v4.17.0.json\", \"datasets\": {\"data-1b001587c028498e70538ed310063e51\": [{\"Word\": \"Iraq\", \"Weight\": -0.9358529031331603}, {\"Word\": \"was\", \"Weight\": -0.0358845002692577}, {\"Word\": \"in\", \"Weight\": -0.017416213388210394}, {\"Word\": \"filmed\", \"Weight\": 0.00802450706528586}, {\"Word\": \"film\", \"Weight\": 0.0077573875142285895}, {\"Word\": \"This\", \"Weight\": 0.0031263867499817305}]}}, {\"mode\": \"vega-lite\"});\n",
"</script>"
],
"text/plain": [
"alt.LayerChart(...)"
]
},
"execution_count": 9,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"single_nearest = alt.selection_single(on='mouseover', nearest=True)\n",
"viz = alt.Chart(limedf).encode(\n",
" alt.X('Weight:Q', scale=alt.Scale(domain=(-1, 1))),\n",
" alt.Y('Word:N', sort='x', axis=None),\n",
" color=alt.Color(\"Weight\", scale=alt.Scale(scheme='blueorange', domain=[0], type=\"threshold\", range='diverging'), legend=None),\n",
" tooltip = (\"Word\",\"Weight\")\n",
").mark_bar().properties(title =\"Importance of individual words\")\n",
"\n",
"text = viz.mark_text(\n",
" fill=\"black\",\n",
" align='right',\n",
" baseline='middle'\n",
").encode(\n",
" text='Word:N'\n",
")\n",
"limeplot = alt.LayerChart(layer=[viz,text], width = 300).configure_axis(grid=False).configure_view(strokeWidth=0)\n",
"limeplot"
]
},
{
"cell_type": "markdown",
"id": "bf0512b6-336e-4842-9bde-34e03a1ca7c6",
"metadata": {},
"source": [
"### Testing predictions and visualization\n",
"Here I will attempt to import the model from huggingface, generate predictions for each of the sentences, and then visualize those predictions into a dot plot. If I can get this to work then I will move on to testing a full pipeline for letting the user pick which part of the sentence they wish to generate counterfactuals for."
]
},
{
"cell_type": "code",
"execution_count": 10,
"id": "74c639bb-e74a-4a46-8047-3552265ae6a4",
"metadata": {},
"outputs": [],
"source": [
"#Discovering that there's a pipeline specifically to provide scores. \n",
"#I used it to get a list of lists of dictionaries that I can then manipulate to calculate the proper prediction score.\n",
"pipe = TextClassificationPipeline(model=model, tokenizer=tokenizer, return_all_scores=True)"
]
},
{
"cell_type": "code",
"execution_count": 11,
"id": "8726a284-99bd-47f1-9756-1c3ae603db10",
"metadata": {},
"outputs": [],
"source": [
"def eval_pred(text):\n",
" '''A basic function for evaluating the prediction from the model and turning it into a visualization friendly number.'''\n",
" preds = pipe(text)\n",
" neg_score = preds[0][0]['score']\n",
" pos_score = preds[0][1]['score']\n",
" if pos_score >= neg_score:\n",
" return pos_score\n",
" if neg_score >= pos_score:\n",
" return -1 * neg_score"
]
},
{
"cell_type": "code",
"execution_count": 12,
"id": "f38f5061-f30a-4c81-9465-37951c3ad9f4",
"metadata": {},
"outputs": [],
"source": [
"def eval_pred_test(text, return_all = False):\n",
" '''A basic function for evaluating the prediction from the model and turning it into a visualization friendly number.'''\n",
" preds = pipe(text)\n",
" neg_score = -1 * preds[0][0]['score']\n",
" sent_neg = preds[0][0]['label']\n",
" pos_score = preds[0][1]['score']\n",
" sent_pos = preds[0][1]['label']\n",
" prediction = 0\n",
" sentiment = ''\n",
" if pos_score > abs(neg_score):\n",
" prediction = pos_score\n",
" sentiment = sent_pos\n",
" elif abs(neg_score) > pos_score:\n",
" prediction = neg_score\n",
" sentiment = sent_neg\n",
" \n",
" if return_all:\n",
" return prediction, sentiment\n",
" else:\n",
" return prediction"
]
},
{
"cell_type": "markdown",
"id": "8b349a87-fe83-4045-a63a-d054489bb461",
"metadata": {},
"source": [
"## Load the dummy countries I created to test generating counterfactuals\n",
"I decided to test the pipeline with a known problem space. Taking the text from Aurélien Géron's observations in twitter, I built a built a small scale test using the learnings I had to prove that we can identify a particular part of speech, use it to generate counterfactuals, and then build a visualization off it."
]
},
{
"cell_type": "code",
"execution_count": 13,
"id": "46ab3332-964c-449f-8cef-a9ff7df397a4",
"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>Country</th>\n",
" <th>Continent</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>Algeria</td>\n",
" <td>Africa</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>Angola</td>\n",
" <td>Africa</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>Benin</td>\n",
" <td>Africa</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>Botswana</td>\n",
" <td>Africa</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>Burkina</td>\n",
" <td>Africa</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Country Continent\n",
"0 Algeria Africa\n",
"1 Angola Africa\n",
"2 Benin Africa\n",
"3 Botswana Africa\n",
"4 Burkina Africa"
]
},
"execution_count": 13,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"#load my test data from https://github.com/dbouquin/IS_608/blob/master/NanosatDB_munging/Countries-Continents.csv\n",
"df = pd.read_csv(\"Assets/Countries/countries.csv\")\n",
"df.head()"
]
},
{
"cell_type": "code",
"execution_count": 14,
"id": "51c75894-80af-4625-8ce8-660e500b496b",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"From This film, ['film'] added to pos_options due to wildcard.\n",
"From Iraq, ['Iraq'] and ['film', 'Iraq'] added to pos_options due to entity recognition.\n",
"['film', 'Iraq']\n"
]
},
{
"data": {
"text/plain": [
"'Iraq'"
]
},
"execution_count": 14,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"#Note: we will need to build the function that lets the user choose from the options available. For now I have hard coded it as \"selection\", from \"user_options\".\n",
"user_options = select_crit(doc4)\n",
"print(user_options)\n",
"selection = user_options[1]\n",
"selection"
]
},
{
"cell_type": "code",
"execution_count": 15,
"id": "3d6419f1-bf7d-44bc-afb8-ac26ef9002df",
"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>Country</th>\n",
" <th>Continent</th>\n",
" <th>text</th>\n",
" <th>prediction</th>\n",
" <th>seed</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>Algeria</td>\n",
" <td>Africa</td>\n",
" <td>This film was filmed in Algeria.</td>\n",
" <td>0.806454</td>\n",
" <td>alternative</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>Angola</td>\n",
" <td>Africa</td>\n",
" <td>This film was filmed in Angola.</td>\n",
" <td>-0.775854</td>\n",
" <td>alternative</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>Benin</td>\n",
" <td>Africa</td>\n",
" <td>This film was filmed in Benin.</td>\n",
" <td>0.962272</td>\n",
" <td>alternative</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>Botswana</td>\n",
" <td>Africa</td>\n",
" <td>This film was filmed in Botswana.</td>\n",
" <td>0.785837</td>\n",
" <td>alternative</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>Burkina</td>\n",
" <td>Africa</td>\n",
" <td>This film was filmed in Burkina.</td>\n",
" <td>0.872980</td>\n",
" <td>alternative</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Country Continent text prediction \\\n",
"0 Algeria Africa This film was filmed in Algeria. 0.806454 \n",
"1 Angola Africa This film was filmed in Angola. -0.775854 \n",
"2 Benin Africa This film was filmed in Benin. 0.962272 \n",
"3 Botswana Africa This film was filmed in Botswana. 0.785837 \n",
"4 Burkina Africa This film was filmed in Burkina. 0.872980 \n",
"\n",
" seed \n",
"0 alternative \n",
"1 alternative \n",
"2 alternative \n",
"3 alternative \n",
"4 alternative "
]
},
"execution_count": 15,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"#Create a function that generates the counterfactuals within a data frame.\n",
"def gen_cf_country(df,document,selection):\n",
" df['text'] = df.Country.apply(lambda x: document.text.replace(selection,x))\n",
" df['prediction'] = df.text.apply(eval_pred_test)\n",
" #added this because I think it will make the end results better if we ensure the seed is in the data we generate counterfactuals from.\n",
" df['seed'] = df.Country.apply(lambda x: 'seed' if x == selection else 'alternative')\n",
" return df\n",
"\n",
"df = gen_cf_country(df,doc4,selection)\n",
"df.head()"
]
},
{
"cell_type": "code",
"execution_count": 16,
"id": "ecb9dd41-2fab-49bd-bae5-30300ce39e41",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"<div id=\"altair-viz-b04081e2f48148ebbc743fff61e76f2f\"></div>\n",
"<script type=\"text/javascript\">\n",
" var VEGA_DEBUG = (typeof VEGA_DEBUG == \"undefined\") ? {} : VEGA_DEBUG;\n",
" (function(spec, embedOpt){\n",
" let outputDiv = document.currentScript.previousElementSibling;\n",
" if (outputDiv.id !== \"altair-viz-b04081e2f48148ebbc743fff61e76f2f\") {\n",
" outputDiv = document.getElementById(\"altair-viz-b04081e2f48148ebbc743fff61e76f2f\");\n",
" }\n",
" const paths = {\n",
" \"vega\": \"https://cdn.jsdelivr.net/npm//vega@5?noext\",\n",
" \"vega-lib\": \"https://cdn.jsdelivr.net/npm//vega-lib?noext\",\n",
" \"vega-lite\": \"https://cdn.jsdelivr.net/npm//vega-lite@4.17.0?noext\",\n",
" \"vega-embed\": \"https://cdn.jsdelivr.net/npm//vega-embed@6?noext\",\n",
" };\n",
"\n",
" function maybeLoadScript(lib, version) {\n",
" var key = `${lib.replace(\"-\", \"\")}_version`;\n",
" return (VEGA_DEBUG[key] == version) ?\n",
" Promise.resolve(paths[lib]) :\n",
" new Promise(function(resolve, reject) {\n",
" var s = document.createElement('script');\n",
" document.getElementsByTagName(\"head\")[0].appendChild(s);\n",
" s.async = true;\n",
" s.onload = () => {\n",
" VEGA_DEBUG[key] = version;\n",
" return resolve(paths[lib]);\n",
" };\n",
" s.onerror = () => reject(`Error loading script: ${paths[lib]}`);\n",
" s.src = paths[lib];\n",
" });\n",
" }\n",
"\n",
" function showError(err) {\n",
" outputDiv.innerHTML = `<div class=\"error\" style=\"color:red;\">${err}</div>`;\n",
" throw err;\n",
" }\n",
"\n",
" function displayChart(vegaEmbed) {\n",
" vegaEmbed(outputDiv, spec, embedOpt)\n",
" .catch(err => showError(`Javascript Error: ${err.message}<br>This usually means there's a typo in your chart specification. See the javascript console for the full traceback.`));\n",
" }\n",
"\n",
" if(typeof define === \"function\" && define.amd) {\n",
" requirejs.config({paths});\n",
" require([\"vega-embed\"], displayChart, err => showError(`Error loading script: ${err.message}`));\n",
" } else {\n",
" maybeLoadScript(\"vega\", \"5\")\n",
" .then(() => maybeLoadScript(\"vega-lite\", \"4.17.0\"))\n",
" .then(() => maybeLoadScript(\"vega-embed\", \"6\"))\n",
" .catch(showError)\n",
" .then(() => displayChart(vegaEmbed));\n",
" }\n",
" })({\"config\": {\"view\": {\"continuousWidth\": 400, \"continuousHeight\": 300}}, \"data\": {\"name\": \"data-d6144c20ed1c104065f226d393d7e424\"}, \"mark\": {\"type\": \"circle\", \"opacity\": 0.5}, \"encoding\": {\"color\": {\"field\": \"seed\", \"legend\": {\"title\": \"Seed or Alternative\"}, \"type\": \"nominal\"}, \"size\": {\"field\": \"seed\", \"type\": \"nominal\"}, \"tooltip\": [{\"field\": \"Country\", \"type\": \"nominal\"}, {\"field\": \"prediction\", \"type\": \"quantitative\"}], \"x\": {\"field\": \"Continent\", \"type\": \"nominal\"}, \"y\": {\"field\": \"prediction\", \"type\": \"quantitative\"}}, \"selection\": {\"selector002\": {\"type\": \"single\", \"on\": \"mouseover\", \"nearest\": true}}, \"width\": 300, \"$schema\": \"https://vega.github.io/schema/vega-lite/v4.17.0.json\", \"datasets\": {\"data-d6144c20ed1c104065f226d393d7e424\": [{\"Country\": \"Algeria\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Algeria.\", \"prediction\": 0.8064541816711426, \"seed\": \"alternative\"}, {\"Country\": \"Angola\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Angola.\", \"prediction\": -0.7758541703224182, \"seed\": \"alternative\"}, {\"Country\": \"Benin\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Benin.\", \"prediction\": 0.9622722268104553, \"seed\": \"alternative\"}, {\"Country\": \"Botswana\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Botswana.\", \"prediction\": 0.7858365774154663, \"seed\": \"alternative\"}, {\"Country\": \"Burkina\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Burkina.\", \"prediction\": 0.8729804754257202, \"seed\": \"alternative\"}, {\"Country\": \"Burundi\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Burundi.\", \"prediction\": -0.6306232810020447, \"seed\": \"alternative\"}, {\"Country\": \"Cameroon\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Cameroon.\", \"prediction\": 0.5283073782920837, \"seed\": \"alternative\"}, {\"Country\": \"Cape Verde\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Cape Verde.\", \"prediction\": 0.8932027220726013, \"seed\": \"alternative\"}, {\"Country\": \"Central African Republic\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Central African Republic.\", \"prediction\": 0.9326885342597961, \"seed\": \"alternative\"}, {\"Country\": \"Chad\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Chad.\", \"prediction\": 0.788737952709198, \"seed\": \"alternative\"}, {\"Country\": \"Comoros\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Comoros.\", \"prediction\": 0.9623100757598877, \"seed\": \"alternative\"}, {\"Country\": \"Congo\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Congo.\", \"prediction\": 0.6309685707092285, \"seed\": \"alternative\"}, {\"Country\": \"Congo, Democratic Republic of\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Congo, Democratic Republic of.\", \"prediction\": -0.54060298204422, \"seed\": \"alternative\"}, {\"Country\": \"Djibouti\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Djibouti.\", \"prediction\": 0.8894529938697815, \"seed\": \"alternative\"}, {\"Country\": \"Egypt\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Egypt.\", \"prediction\": 0.9648140072822571, \"seed\": \"alternative\"}, {\"Country\": \"Equatorial Guinea\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Equatorial Guinea.\", \"prediction\": 0.6021467447280884, \"seed\": \"alternative\"}, {\"Country\": \"Eritrea\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Eritrea.\", \"prediction\": 0.5404142141342163, \"seed\": \"alternative\"}, {\"Country\": \"Ethiopia\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Ethiopia.\", \"prediction\": 0.7997546195983887, \"seed\": \"alternative\"}, {\"Country\": \"Gabon\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Gabon.\", \"prediction\": -0.8517823219299316, \"seed\": \"alternative\"}, {\"Country\": \"Gambia\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Gambia.\", \"prediction\": -0.5401656031608582, \"seed\": \"alternative\"}, {\"Country\": \"Ghana\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Ghana.\", \"prediction\": 0.9684805870056152, \"seed\": \"alternative\"}, {\"Country\": \"Guinea\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Guinea.\", \"prediction\": 0.6188081502914429, \"seed\": \"alternative\"}, {\"Country\": \"Guinea-Bissau\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Guinea-Bissau.\", \"prediction\": -0.500963032245636, \"seed\": \"alternative\"}, {\"Country\": \"Ivory Coast\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Ivory Coast.\", \"prediction\": 0.9872506856918335, \"seed\": \"alternative\"}, {\"Country\": \"Kenya\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Kenya.\", \"prediction\": 0.9789031744003296, \"seed\": \"alternative\"}, {\"Country\": \"Lesotho\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Lesotho.\", \"prediction\": 0.6674107313156128, \"seed\": \"alternative\"}, {\"Country\": \"Liberia\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Liberia.\", \"prediction\": -0.6720185279846191, \"seed\": \"alternative\"}, {\"Country\": \"Libya\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Libya.\", \"prediction\": 0.53217613697052, \"seed\": \"alternative\"}, {\"Country\": \"Madagascar\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Madagascar.\", \"prediction\": 0.9730344414710999, \"seed\": \"alternative\"}, {\"Country\": \"Malawi\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Malawi.\", \"prediction\": -0.7816339135169983, \"seed\": \"alternative\"}, {\"Country\": \"Mali\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Mali.\", \"prediction\": -0.6651991009712219, \"seed\": \"alternative\"}, {\"Country\": \"Mauritania\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Mauritania.\", \"prediction\": 0.6149344444274902, \"seed\": \"alternative\"}, {\"Country\": \"Mauritius\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Mauritius.\", \"prediction\": 0.9310740828514099, \"seed\": \"alternative\"}, {\"Country\": \"Morocco\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Morocco.\", \"prediction\": 0.9121577143669128, \"seed\": \"alternative\"}, {\"Country\": \"Mozambique\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Mozambique.\", \"prediction\": -0.7047757506370544, \"seed\": \"alternative\"}, {\"Country\": \"Namibia\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Namibia.\", \"prediction\": -0.5836523175239563, \"seed\": \"alternative\"}, {\"Country\": \"Niger\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Niger.\", \"prediction\": -0.6313472390174866, \"seed\": \"alternative\"}, {\"Country\": \"Nigeria\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Nigeria.\", \"prediction\": 0.7361583113670349, \"seed\": \"alternative\"}, {\"Country\": \"Rwanda\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Rwanda.\", \"prediction\": -0.7642565965652466, \"seed\": \"alternative\"}, {\"Country\": \"Sao Tome and Principe\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Sao Tome and Principe.\", \"prediction\": 0.6587044596672058, \"seed\": \"alternative\"}, {\"Country\": \"Senegal\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Senegal.\", \"prediction\": 0.8155898451805115, \"seed\": \"alternative\"}, {\"Country\": \"Seychelles\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Seychelles.\", \"prediction\": 0.8802894949913025, \"seed\": \"alternative\"}, {\"Country\": \"Sierra Leone\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Sierra Leone.\", \"prediction\": 0.9483919143676758, \"seed\": \"alternative\"}, {\"Country\": \"Somalia\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Somalia.\", \"prediction\": -0.6477505564689636, \"seed\": \"alternative\"}, {\"Country\": \"South Africa\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in South Africa.\", \"prediction\": 0.5048943161964417, \"seed\": \"alternative\"}, {\"Country\": \"South Sudan\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in South Sudan.\", \"prediction\": -0.8506219983100891, \"seed\": \"alternative\"}, {\"Country\": \"Sudan\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Sudan.\", \"prediction\": -0.8910807967185974, \"seed\": \"alternative\"}, {\"Country\": \"Swaziland\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Swaziland.\", \"prediction\": 0.7761040925979614, \"seed\": \"alternative\"}, {\"Country\": \"Tanzania\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Tanzania.\", \"prediction\": 0.669053316116333, \"seed\": \"alternative\"}, {\"Country\": \"Togo\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Togo.\", \"prediction\": 0.9404287934303284, \"seed\": \"alternative\"}, {\"Country\": \"Tunisia\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Tunisia.\", \"prediction\": 0.8345948457717896, \"seed\": \"alternative\"}, {\"Country\": \"Uganda\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Uganda.\", \"prediction\": 0.7823328971862793, \"seed\": \"alternative\"}, {\"Country\": \"Zambia\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Zambia.\", \"prediction\": -0.6479448080062866, \"seed\": \"alternative\"}, {\"Country\": \"Zimbabwe\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Zimbabwe.\", \"prediction\": 0.7163158059120178, \"seed\": \"alternative\"}, {\"Country\": \"Afghanistan\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Afghanistan.\", \"prediction\": -0.8350331783294678, \"seed\": \"alternative\"}, {\"Country\": \"Bahrain\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Bahrain.\", \"prediction\": 0.9627965092658997, \"seed\": \"alternative\"}, {\"Country\": \"Bangladesh\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Bangladesh.\", \"prediction\": 0.6659616231918335, \"seed\": \"alternative\"}, {\"Country\": \"Bhutan\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Bhutan.\", \"prediction\": 0.9108285307884216, \"seed\": \"alternative\"}, {\"Country\": \"Brunei\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Brunei.\", \"prediction\": 0.7673805952072144, \"seed\": \"alternative\"}, {\"Country\": \"Burma (Myanmar)\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Burma (Myanmar).\", \"prediction\": 0.5261574387550354, \"seed\": \"alternative\"}, {\"Country\": \"Cambodia\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Cambodia.\", \"prediction\": 0.9706045389175415, \"seed\": \"alternative\"}, {\"Country\": \"China\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in China.\", \"prediction\": 0.6985915303230286, \"seed\": \"alternative\"}, {\"Country\": \"East Timor\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in East Timor.\", \"prediction\": -0.7553014159202576, \"seed\": \"alternative\"}, {\"Country\": \"India\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in India.\", \"prediction\": 0.9856906533241272, \"seed\": \"alternative\"}, {\"Country\": \"Indonesia\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Indonesia.\", \"prediction\": 0.9617947936058044, \"seed\": \"alternative\"}, {\"Country\": \"Iran\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Iran.\", \"prediction\": 0.935718834400177, \"seed\": \"alternative\"}, {\"Country\": \"Iraq\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Iraq.\", \"prediction\": -0.9768388867378235, \"seed\": \"seed\"}, {\"Country\": \"Israel\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Israel.\", \"prediction\": 0.8940765261650085, \"seed\": \"alternative\"}, {\"Country\": \"Japan\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Japan.\", \"prediction\": 0.8561221957206726, \"seed\": \"alternative\"}, {\"Country\": \"Jordan\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Jordan.\", \"prediction\": 0.5632433891296387, \"seed\": \"alternative\"}, {\"Country\": \"Kazakhstan\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Kazakhstan.\", \"prediction\": 0.8813521862030029, \"seed\": \"alternative\"}, {\"Country\": \"Korea, North\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Korea, North.\", \"prediction\": -0.692742645740509, \"seed\": \"alternative\"}, {\"Country\": \"Korea, South\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Korea, South.\", \"prediction\": 0.7591306567192078, \"seed\": \"alternative\"}, {\"Country\": \"Kuwait\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Kuwait.\", \"prediction\": 0.9136238098144531, \"seed\": \"alternative\"}, {\"Country\": \"Kyrgyzstan\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Kyrgyzstan.\", \"prediction\": 0.9416173100471497, \"seed\": \"alternative\"}, {\"Country\": \"Laos\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Laos.\", \"prediction\": 0.7455804347991943, \"seed\": \"alternative\"}, {\"Country\": \"Lebanon\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Lebanon.\", \"prediction\": 0.9018603563308716, \"seed\": \"alternative\"}, {\"Country\": \"Malaysia\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Malaysia.\", \"prediction\": 0.9053533673286438, \"seed\": \"alternative\"}, {\"Country\": \"Maldives\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Maldives.\", \"prediction\": 0.8150556087493896, \"seed\": \"alternative\"}, {\"Country\": \"Mongolia\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Mongolia.\", \"prediction\": 0.9706059098243713, \"seed\": \"alternative\"}, {\"Country\": \"Nepal\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Nepal.\", \"prediction\": 0.9837730526924133, \"seed\": \"alternative\"}, {\"Country\": \"Oman\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Oman.\", \"prediction\": 0.8641175627708435, \"seed\": \"alternative\"}, {\"Country\": \"Pakistan\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Pakistan.\", \"prediction\": 0.8881147503852844, \"seed\": \"alternative\"}, {\"Country\": \"Philippines\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Philippines.\", \"prediction\": 0.9892238974571228, \"seed\": \"alternative\"}, {\"Country\": \"Qatar\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Qatar.\", \"prediction\": 0.9696690440177917, \"seed\": \"alternative\"}, {\"Country\": \"Russian Federation\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Russian Federation.\", \"prediction\": 0.9777944087982178, \"seed\": \"alternative\"}, {\"Country\": \"Saudi Arabia\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Saudi Arabia.\", \"prediction\": -0.7760475873947144, \"seed\": \"alternative\"}, {\"Country\": \"Singapore\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Singapore.\", \"prediction\": 0.9684174060821533, \"seed\": \"alternative\"}, {\"Country\": \"Sri Lanka\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Sri Lanka.\", \"prediction\": 0.9552921056747437, \"seed\": \"alternative\"}, {\"Country\": \"Syria\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Syria.\", \"prediction\": -0.8887014985084534, \"seed\": \"alternative\"}, {\"Country\": \"Tajikistan\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Tajikistan.\", \"prediction\": 0.8012317419052124, \"seed\": \"alternative\"}, {\"Country\": \"Thailand\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Thailand.\", \"prediction\": 0.8334980607032776, \"seed\": \"alternative\"}, {\"Country\": \"Turkey\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Turkey.\", \"prediction\": 0.5693907141685486, \"seed\": \"alternative\"}, {\"Country\": \"Turkmenistan\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Turkmenistan.\", \"prediction\": 0.8194981813430786, \"seed\": \"alternative\"}, {\"Country\": \"United Arab Emirates\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in United Arab Emirates.\", \"prediction\": 0.921615719795227, \"seed\": \"alternative\"}, {\"Country\": \"Uzbekistan\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Uzbekistan.\", \"prediction\": 0.8483680486679077, \"seed\": \"alternative\"}, {\"Country\": \"Vietnam\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Vietnam.\", \"prediction\": -0.9427406191825867, \"seed\": \"alternative\"}, {\"Country\": \"Yemen\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Yemen.\", \"prediction\": -0.8567103743553162, \"seed\": \"alternative\"}, {\"Country\": \"Albania\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Albania.\", \"prediction\": 0.9874222278594971, \"seed\": \"alternative\"}, {\"Country\": \"Andorra\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Andorra.\", \"prediction\": 0.9597309231758118, \"seed\": \"alternative\"}, {\"Country\": \"Armenia\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Armenia.\", \"prediction\": 0.986950695514679, \"seed\": \"alternative\"}, {\"Country\": \"Austria\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Austria.\", \"prediction\": 0.8858200907707214, \"seed\": \"alternative\"}, {\"Country\": \"Azerbaijan\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Azerbaijan.\", \"prediction\": 0.9770861268043518, \"seed\": \"alternative\"}, {\"Country\": \"Belarus\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Belarus.\", \"prediction\": 0.5220555663108826, \"seed\": \"alternative\"}, {\"Country\": \"Belgium\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Belgium.\", \"prediction\": 0.9663146138191223, \"seed\": \"alternative\"}, {\"Country\": \"Bosnia and Herzegovina\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Bosnia and Herzegovina.\", \"prediction\": 0.9699962139129639, \"seed\": \"alternative\"}, {\"Country\": \"Bulgaria\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Bulgaria.\", \"prediction\": 0.8968954086303711, \"seed\": \"alternative\"}, {\"Country\": \"Croatia\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Croatia.\", \"prediction\": 0.8545156717300415, \"seed\": \"alternative\"}, {\"Country\": \"Cyprus\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Cyprus.\", \"prediction\": 0.9457007646560669, \"seed\": \"alternative\"}, {\"Country\": \"CZ\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in CZ.\", \"prediction\": -0.9620359539985657, \"seed\": \"alternative\"}, {\"Country\": \"Denmark\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Denmark.\", \"prediction\": 0.9433714747428894, \"seed\": \"alternative\"}, {\"Country\": \"Estonia\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Estonia.\", \"prediction\": 0.9754448533058167, \"seed\": \"alternative\"}, {\"Country\": \"Finland\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Finland.\", \"prediction\": 0.9832987189292908, \"seed\": \"alternative\"}, {\"Country\": \"France\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in France.\", \"prediction\": 0.9652075171470642, \"seed\": \"alternative\"}, {\"Country\": \"Georgia\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Georgia.\", \"prediction\": 0.9579687714576721, \"seed\": \"alternative\"}, {\"Country\": \"Germany\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Germany.\", \"prediction\": -0.7719752192497253, \"seed\": \"alternative\"}, {\"Country\": \"Greece\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Greece.\", \"prediction\": 0.974821925163269, \"seed\": \"alternative\"}, {\"Country\": \"Hungary\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Hungary.\", \"prediction\": 0.9794204831123352, \"seed\": \"alternative\"}, {\"Country\": \"Iceland\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Iceland.\", \"prediction\": 0.9596456289291382, \"seed\": \"alternative\"}, {\"Country\": \"Ireland\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Ireland.\", \"prediction\": 0.9691770076751709, \"seed\": \"alternative\"}, {\"Country\": \"Italy\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Italy.\", \"prediction\": 0.973678469657898, \"seed\": \"alternative\"}, {\"Country\": \"Latvia\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Latvia.\", \"prediction\": 0.9340384006500244, \"seed\": \"alternative\"}, {\"Country\": \"Liechtenstein\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Liechtenstein.\", \"prediction\": 0.9714267253875732, \"seed\": \"alternative\"}, {\"Country\": \"Lithuania\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Lithuania.\", \"prediction\": 0.9562608599662781, \"seed\": \"alternative\"}, {\"Country\": \"Luxembourg\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Luxembourg.\", \"prediction\": 0.9322720170021057, \"seed\": \"alternative\"}, {\"Country\": \"Macedonia\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Macedonia.\", \"prediction\": 0.8895869255065918, \"seed\": \"alternative\"}, {\"Country\": \"Malta\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Malta.\", \"prediction\": 0.979903519153595, \"seed\": \"alternative\"}, {\"Country\": \"Moldova\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Moldova.\", \"prediction\": 0.8919235467910767, \"seed\": \"alternative\"}, {\"Country\": \"Monaco\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Monaco.\", \"prediction\": 0.9971835017204285, \"seed\": \"alternative\"}, {\"Country\": \"Montenegro\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Montenegro.\", \"prediction\": 0.9382426738739014, \"seed\": \"alternative\"}, {\"Country\": \"Netherlands\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Netherlands.\", \"prediction\": 0.9562605023384094, \"seed\": \"alternative\"}, {\"Country\": \"Norway\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Norway.\", \"prediction\": 0.9528943300247192, \"seed\": \"alternative\"}, {\"Country\": \"Poland\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Poland.\", \"prediction\": 0.9124379754066467, \"seed\": \"alternative\"}, {\"Country\": \"Portugal\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Portugal.\", \"prediction\": 0.9363807439804077, \"seed\": \"alternative\"}, {\"Country\": \"Romania\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Romania.\", \"prediction\": 0.982775866985321, \"seed\": \"alternative\"}, {\"Country\": \"San Marino\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in San Marino.\", \"prediction\": 0.924018144607544, \"seed\": \"alternative\"}, {\"Country\": \"Serbia\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Serbia.\", \"prediction\": 0.740748405456543, \"seed\": \"alternative\"}, {\"Country\": \"Slovakia\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Slovakia.\", \"prediction\": 0.5953425168991089, \"seed\": \"alternative\"}, {\"Country\": \"Slovenia\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Slovenia.\", \"prediction\": 0.8840153217315674, \"seed\": \"alternative\"}, {\"Country\": \"Spain\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Spain.\", \"prediction\": 0.9535741209983826, \"seed\": \"alternative\"}, {\"Country\": \"Sweden\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Sweden.\", \"prediction\": 0.9694980382919312, \"seed\": \"alternative\"}, {\"Country\": \"Switzerland\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Switzerland.\", \"prediction\": 0.7584144473075867, \"seed\": \"alternative\"}, {\"Country\": \"Ukraine\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Ukraine.\", \"prediction\": 0.7340573668479919, \"seed\": \"alternative\"}, {\"Country\": \"United Kingdom\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in United Kingdom.\", \"prediction\": 0.8982904553413391, \"seed\": \"alternative\"}, {\"Country\": \"Vatican City\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Vatican City.\", \"prediction\": 0.7796335816383362, \"seed\": \"alternative\"}, {\"Country\": \"Antigua and Barbuda\", \"Continent\": \"North America\", \"text\": \"This film was filmed in Antigua and Barbuda.\", \"prediction\": 0.9056354761123657, \"seed\": \"alternative\"}, {\"Country\": \"Bahamas\", \"Continent\": \"North America\", \"text\": \"This film was filmed in Bahamas.\", \"prediction\": 0.9206929802894592, \"seed\": \"alternative\"}, {\"Country\": \"Barbados\", \"Continent\": \"North America\", \"text\": \"This film was filmed in Barbados.\", \"prediction\": 0.9170283079147339, \"seed\": \"alternative\"}, {\"Country\": \"Belize\", \"Continent\": \"North America\", \"text\": \"This film was filmed in Belize.\", \"prediction\": 0.9203323125839233, \"seed\": \"alternative\"}, {\"Country\": \"Canada\", \"Continent\": \"North America\", \"text\": \"This film was filmed in Canada.\", \"prediction\": 0.9400970339775085, \"seed\": \"alternative\"}, {\"Country\": \"Costa Rica\", \"Continent\": \"North America\", \"text\": \"This film was filmed in Costa Rica.\", \"prediction\": 0.9815211892127991, \"seed\": \"alternative\"}, {\"Country\": \"Cuba\", \"Continent\": \"North America\", \"text\": \"This film was filmed in Cuba.\", \"prediction\": 0.7347409725189209, \"seed\": \"alternative\"}, {\"Country\": \"Dominica\", \"Continent\": \"North America\", \"text\": \"This film was filmed in Dominica.\", \"prediction\": 0.5335615277290344, \"seed\": \"alternative\"}, {\"Country\": \"Dominican Republic\", \"Continent\": \"North America\", \"text\": \"This film was filmed in Dominican Republic.\", \"prediction\": 0.9594704508781433, \"seed\": \"alternative\"}, {\"Country\": \"El Salvador\", \"Continent\": \"North America\", \"text\": \"This film was filmed in El Salvador.\", \"prediction\": 0.9804539084434509, \"seed\": \"alternative\"}, {\"Country\": \"Grenada\", \"Continent\": \"North America\", \"text\": \"This film was filmed in Grenada.\", \"prediction\": 0.6266372799873352, \"seed\": \"alternative\"}, {\"Country\": \"Guatemala\", \"Continent\": \"North America\", \"text\": \"This film was filmed in Guatemala.\", \"prediction\": 0.7368012070655823, \"seed\": \"alternative\"}, {\"Country\": \"Haiti\", \"Continent\": \"North America\", \"text\": \"This film was filmed in Haiti.\", \"prediction\": 0.9208669662475586, \"seed\": \"alternative\"}, {\"Country\": \"Honduras\", \"Continent\": \"North America\", \"text\": \"This film was filmed in Honduras.\", \"prediction\": 0.7440645098686218, \"seed\": \"alternative\"}, {\"Country\": \"Jamaica\", \"Continent\": \"North America\", \"text\": \"This film was filmed in Jamaica.\", \"prediction\": 0.8702073097229004, \"seed\": \"alternative\"}, {\"Country\": \"Mexico\", \"Continent\": \"North America\", \"text\": \"This film was filmed in Mexico.\", \"prediction\": 0.9770798683166504, \"seed\": \"alternative\"}, {\"Country\": \"Nicaragua\", \"Continent\": \"North America\", \"text\": \"This film was filmed in Nicaragua.\", \"prediction\": -0.6681438684463501, \"seed\": \"alternative\"}, {\"Country\": \"Panama\", \"Continent\": \"North America\", \"text\": \"This film was filmed in Panama.\", \"prediction\": 0.737115740776062, \"seed\": \"alternative\"}, {\"Country\": \"Saint Kitts and Nevis\", \"Continent\": \"North America\", \"text\": \"This film was filmed in Saint Kitts and Nevis.\", \"prediction\": 0.9829047918319702, \"seed\": \"alternative\"}, {\"Country\": \"Saint Lucia\", \"Continent\": \"North America\", \"text\": \"This film was filmed in Saint Lucia.\", \"prediction\": 0.7933508157730103, \"seed\": \"alternative\"}, {\"Country\": \"Saint Vincent and the Grenadines\", \"Continent\": \"North America\", \"text\": \"This film was filmed in Saint Vincent and the Grenadines.\", \"prediction\": 0.8782792091369629, \"seed\": \"alternative\"}, {\"Country\": \"Trinidad and Tobago\", \"Continent\": \"North America\", \"text\": \"This film was filmed in Trinidad and Tobago.\", \"prediction\": 0.9884806871414185, \"seed\": \"alternative\"}, {\"Country\": \"US\", \"Continent\": \"North America\", \"text\": \"This film was filmed in US.\", \"prediction\": 0.926520586013794, \"seed\": \"alternative\"}, {\"Country\": \"Australia\", \"Continent\": \"Oceania\", \"text\": \"This film was filmed in Australia.\", \"prediction\": 0.9371141195297241, \"seed\": \"alternative\"}, {\"Country\": \"Fiji\", \"Continent\": \"Oceania\", \"text\": \"This film was filmed in Fiji.\", \"prediction\": 0.9061108827590942, \"seed\": \"alternative\"}, {\"Country\": \"Kiribati\", \"Continent\": \"Oceania\", \"text\": \"This film was filmed in Kiribati.\", \"prediction\": 0.9559115767478943, \"seed\": \"alternative\"}, {\"Country\": \"Marshall Islands\", \"Continent\": \"Oceania\", \"text\": \"This film was filmed in Marshall Islands.\", \"prediction\": 0.96001136302948, \"seed\": \"alternative\"}, {\"Country\": \"Micronesia\", \"Continent\": \"Oceania\", \"text\": \"This film was filmed in Micronesia.\", \"prediction\": -0.57024085521698, \"seed\": \"alternative\"}, {\"Country\": \"Nauru\", \"Continent\": \"Oceania\", \"text\": \"This film was filmed in Nauru.\", \"prediction\": 0.9323841333389282, \"seed\": \"alternative\"}, {\"Country\": \"New Zealand\", \"Continent\": \"Oceania\", \"text\": \"This film was filmed in New Zealand.\", \"prediction\": 0.9654895663261414, \"seed\": \"alternative\"}, {\"Country\": \"Palau\", \"Continent\": \"Oceania\", \"text\": \"This film was filmed in Palau.\", \"prediction\": 0.7104437351226807, \"seed\": \"alternative\"}, {\"Country\": \"Papua New Guinea\", \"Continent\": \"Oceania\", \"text\": \"This film was filmed in Papua New Guinea.\", \"prediction\": 0.5819137692451477, \"seed\": \"alternative\"}, {\"Country\": \"Samoa\", \"Continent\": \"Oceania\", \"text\": \"This film was filmed in Samoa.\", \"prediction\": 0.9161322712898254, \"seed\": \"alternative\"}, {\"Country\": \"Solomon Islands\", \"Continent\": \"Oceania\", \"text\": \"This film was filmed in Solomon Islands.\", \"prediction\": 0.9441730976104736, \"seed\": \"alternative\"}, {\"Country\": \"Tonga\", \"Continent\": \"Oceania\", \"text\": \"This film was filmed in Tonga.\", \"prediction\": 0.550994873046875, \"seed\": \"alternative\"}, {\"Country\": \"Tuvalu\", \"Continent\": \"Oceania\", \"text\": \"This film was filmed in Tuvalu.\", \"prediction\": 0.9912257790565491, \"seed\": \"alternative\"}, {\"Country\": \"Vanuatu\", \"Continent\": \"Oceania\", \"text\": \"This film was filmed in Vanuatu.\", \"prediction\": 0.9395317435264587, \"seed\": \"alternative\"}, {\"Country\": \"Argentina\", \"Continent\": \"South America\", \"text\": \"This film was filmed in Argentina.\", \"prediction\": 0.9719653129577637, \"seed\": \"alternative\"}, {\"Country\": \"Bolivia\", \"Continent\": \"South America\", \"text\": \"This film was filmed in Bolivia.\", \"prediction\": 0.8009489178657532, \"seed\": \"alternative\"}, {\"Country\": \"Brazil\", \"Continent\": \"South America\", \"text\": \"This film was filmed in Brazil.\", \"prediction\": 0.968963086605072, \"seed\": \"alternative\"}, {\"Country\": \"Chile\", \"Continent\": \"South America\", \"text\": \"This film was filmed in Chile.\", \"prediction\": 0.8917940258979797, \"seed\": \"alternative\"}, {\"Country\": \"Colombia\", \"Continent\": \"South America\", \"text\": \"This film was filmed in Colombia.\", \"prediction\": 0.731931746006012, \"seed\": \"alternative\"}, {\"Country\": \"Ecuador\", \"Continent\": \"South America\", \"text\": \"This film was filmed in Ecuador.\", \"prediction\": 0.845059335231781, \"seed\": \"alternative\"}, {\"Country\": \"Guyana\", \"Continent\": \"South America\", \"text\": \"This film was filmed in Guyana.\", \"prediction\": 0.6705957055091858, \"seed\": \"alternative\"}, {\"Country\": \"Paraguay\", \"Continent\": \"South America\", \"text\": \"This film was filmed in Paraguay.\", \"prediction\": 0.6165609359741211, \"seed\": \"alternative\"}, {\"Country\": \"Peru\", \"Continent\": \"South America\", \"text\": \"This film was filmed in Peru.\", \"prediction\": 0.7860054969787598, \"seed\": \"alternative\"}, {\"Country\": \"Suriname\", \"Continent\": \"South America\", \"text\": \"This film was filmed in Suriname.\", \"prediction\": 0.9488070607185364, \"seed\": \"alternative\"}, {\"Country\": \"Uruguay\", \"Continent\": \"South America\", \"text\": \"This film was filmed in Uruguay.\", \"prediction\": 0.744226336479187, \"seed\": \"alternative\"}, {\"Country\": \"Venezuela\", \"Continent\": \"South America\", \"text\": \"This film was filmed in Venezuela.\", \"prediction\": 0.8343830108642578, \"seed\": \"alternative\"}]}}, {\"mode\": \"vega-lite\"});\n",
"</script>"
],
"text/plain": [
"alt.Chart(...)"
]
},
"execution_count": 16,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"single_nearest = alt.selection_single(on='mouseover', nearest=True)\n",
"full = alt.Chart(df).encode(\n",
" alt.X('Continent:N'), # specify nominal data\n",
" alt.Y('prediction:Q'), # specify quantitative data\n",
" color=alt.Color('seed:N', legend=alt.Legend(title=\"Seed or Alternative\")),\n",
" size='seed:N',\n",
" tooltip=('Country','prediction')\n",
").mark_circle(opacity=.5).properties(width=300).add_selection(single_nearest)\n",
"\n",
"full"
]
},
{
"cell_type": "code",
"execution_count": 17,
"id": "56bc30d7-03a5-43ff-9dfe-878197628305",
"metadata": {},
"outputs": [],
"source": [
"df2 = df.nlargest(5, 'prediction')\n",
"df3 = df.nsmallest(5, 'prediction')\n",
"frames = [df2,df3]\n",
"results = pd.concat(frames)"
]
},
{
"cell_type": "code",
"execution_count": 18,
"id": "1610bb48-c9b9-4bee-bcb5-999886acb9e3",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"<div id=\"altair-viz-948f4471f5ee4ed8bb2720ca7dd085a7\"></div>\n",
"<script type=\"text/javascript\">\n",
" var VEGA_DEBUG = (typeof VEGA_DEBUG == \"undefined\") ? {} : VEGA_DEBUG;\n",
" (function(spec, embedOpt){\n",
" let outputDiv = document.currentScript.previousElementSibling;\n",
" if (outputDiv.id !== \"altair-viz-948f4471f5ee4ed8bb2720ca7dd085a7\") {\n",
" outputDiv = document.getElementById(\"altair-viz-948f4471f5ee4ed8bb2720ca7dd085a7\");\n",
" }\n",
" const paths = {\n",
" \"vega\": \"https://cdn.jsdelivr.net/npm//vega@5?noext\",\n",
" \"vega-lib\": \"https://cdn.jsdelivr.net/npm//vega-lib?noext\",\n",
" \"vega-lite\": \"https://cdn.jsdelivr.net/npm//vega-lite@4.17.0?noext\",\n",
" \"vega-embed\": \"https://cdn.jsdelivr.net/npm//vega-embed@6?noext\",\n",
" };\n",
"\n",
" function maybeLoadScript(lib, version) {\n",
" var key = `${lib.replace(\"-\", \"\")}_version`;\n",
" return (VEGA_DEBUG[key] == version) ?\n",
" Promise.resolve(paths[lib]) :\n",
" new Promise(function(resolve, reject) {\n",
" var s = document.createElement('script');\n",
" document.getElementsByTagName(\"head\")[0].appendChild(s);\n",
" s.async = true;\n",
" s.onload = () => {\n",
" VEGA_DEBUG[key] = version;\n",
" return resolve(paths[lib]);\n",
" };\n",
" s.onerror = () => reject(`Error loading script: ${paths[lib]}`);\n",
" s.src = paths[lib];\n",
" });\n",
" }\n",
"\n",
" function showError(err) {\n",
" outputDiv.innerHTML = `<div class=\"error\" style=\"color:red;\">${err}</div>`;\n",
" throw err;\n",
" }\n",
"\n",
" function displayChart(vegaEmbed) {\n",
" vegaEmbed(outputDiv, spec, embedOpt)\n",
" .catch(err => showError(`Javascript Error: ${err.message}<br>This usually means there's a typo in your chart specification. See the javascript console for the full traceback.`));\n",
" }\n",
"\n",
" if(typeof define === \"function\" && define.amd) {\n",
" requirejs.config({paths});\n",
" require([\"vega-embed\"], displayChart, err => showError(`Error loading script: ${err.message}`));\n",
" } else {\n",
" maybeLoadScript(\"vega\", \"5\")\n",
" .then(() => maybeLoadScript(\"vega-lite\", \"4.17.0\"))\n",
" .then(() => maybeLoadScript(\"vega-embed\", \"6\"))\n",
" .catch(showError)\n",
" .then(() => displayChart(vegaEmbed));\n",
" }\n",
" })({\"config\": {\"view\": {\"continuousWidth\": 400, \"continuousHeight\": 300}}, \"data\": {\"name\": \"data-09f850c452d77d8e274c73526803ae5c\"}, \"mark\": \"circle\", \"encoding\": {\"color\": {\"field\": \"seed\", \"legend\": {\"title\": \"Seed or Alternative\"}, \"type\": \"nominal\"}, \"size\": {\"field\": \"seed\", \"type\": \"nominal\"}, \"tooltip\": [{\"field\": \"Country\", \"type\": \"nominal\"}, {\"field\": \"prediction\", \"type\": \"quantitative\"}], \"x\": {\"field\": \"prediction\", \"type\": \"quantitative\"}, \"y\": {\"field\": \"Country\", \"sort\": \"-x\", \"type\": \"nominal\"}}, \"selection\": {\"selector002\": {\"type\": \"single\", \"on\": \"mouseover\", \"nearest\": true}}, \"width\": 300, \"$schema\": \"https://vega.github.io/schema/vega-lite/v4.17.0.json\", \"datasets\": {\"data-09f850c452d77d8e274c73526803ae5c\": [{\"Country\": \"Monaco\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Monaco.\", \"prediction\": 0.9971835017204285, \"seed\": \"alternative\"}, {\"Country\": \"Tuvalu\", \"Continent\": \"Oceania\", \"text\": \"This film was filmed in Tuvalu.\", \"prediction\": 0.9912257790565491, \"seed\": \"alternative\"}, {\"Country\": \"Philippines\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Philippines.\", \"prediction\": 0.9892238974571228, \"seed\": \"alternative\"}, {\"Country\": \"Trinidad and Tobago\", \"Continent\": \"North America\", \"text\": \"This film was filmed in Trinidad and Tobago.\", \"prediction\": 0.9884806871414185, \"seed\": \"alternative\"}, {\"Country\": \"Albania\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in Albania.\", \"prediction\": 0.9874222278594971, \"seed\": \"alternative\"}, {\"Country\": \"Iraq\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Iraq.\", \"prediction\": -0.9768388867378235, \"seed\": \"seed\"}, {\"Country\": \"CZ\", \"Continent\": \"Europe\", \"text\": \"This film was filmed in CZ.\", \"prediction\": -0.9620359539985657, \"seed\": \"alternative\"}, {\"Country\": \"Vietnam\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Vietnam.\", \"prediction\": -0.9427406191825867, \"seed\": \"alternative\"}, {\"Country\": \"Sudan\", \"Continent\": \"Africa\", \"text\": \"This film was filmed in Sudan.\", \"prediction\": -0.8910807967185974, \"seed\": \"alternative\"}, {\"Country\": \"Syria\", \"Continent\": \"Asia\", \"text\": \"This film was filmed in Syria.\", \"prediction\": -0.8887014985084534, \"seed\": \"alternative\"}]}}, {\"mode\": \"vega-lite\"});\n",
"</script>"
],
"text/plain": [
"alt.Chart(...)"
]
},
"execution_count": 18,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"bar = alt.Chart(results).encode( \n",
" alt.X('prediction:Q'), \n",
" alt.Y('Country:N', sort=\"-x\"),\n",
" color=alt.Color('seed:N', legend=alt.Legend(title=\"Seed or Alternative\")),\n",
" size='seed:N',\n",
" tooltip=('Country','prediction')\n",
").mark_circle().properties(width=300).add_selection(single_nearest)\n",
"\n",
"bar"
]
},
{
"cell_type": "code",
"execution_count": 34,
"id": "96cd0798-5ac5-4ede-8373-e8ed71ab07b3",
"metadata": {},
"outputs": [],
"source": [
"def critical_words(document, options=False):\n",
" '''This function is meant to select the critical part of a sentence. Critical, in this context means\n",
" the part of the sentence that is either: A) a PROPN from the correct entity group; B) an ADJ associated with a NOUN;\n",
" C) a NOUN that represents gender. It also checks this against what the model thinks is important if the user defines \"options\" as \"LIME\" or True.'''\n",
" if type(document) is not spacy.tokens.doc.Doc:\n",
" document = nlp(document)\n",
" chunks = list(document.noun_chunks)\n",
" pos_options = []\n",
" lime_options = []\n",
" \n",
" #Identify what the model cares about.\n",
" if options:\n",
" exp = explainer.explain_instance(document.text, predictor, num_features=15, num_samples=2000)\n",
" lime_results = exp.as_list()\n",
" for feature in lime_results:\n",
" lime_options.append(feature[0])\n",
" lime_results = pd.DataFrame(lime_results, columns=[\"Word\",\"Weight\"])\n",
" \n",
" #Identify what we care about \"parts of speech\". The first section focuses on NOUNs and related ADJ.\n",
" for chunk in chunks:\n",
" #The use of chunk[-1] is due to testing that it appears to always match the root\n",
" root = chunk[-1]\n",
" #This currently matches to a list I've created. I don't know the best way to deal with this so I'm leaving it as is for the moment.\n",
" if root.ent_type_:\n",
" cur_values = []\n",
" if (len(chunk) > 1) and (chunk[-2].dep_ == \"compound\"):\n",
" #creates the compound element of the noun\n",
" compound = [x.text for x in chunk if x.dep_ == \"compound\"]\n",
" print(f\"This is the contents of {compound} and it is {all(elem in lime_options for elem in compound)} that all elements are present in {lime_options}.\") #for QA\n",
" #checks to see all elements in the compound are important to the model or use the compound if not checking importance.\n",
" if (all(elem in lime_options for elem in cur_values) and (options is True)) or ((options is False)):\n",
" #creates a span for the entirety of the compound noun and adds it to the list.\n",
" span = -1 * (1 + len(compound))\n",
" pos_options.append(chunk[span:].text)\n",
" cur_values + [token.text for token in chunk if token.pos_ == \"ADJ\"]\n",
" else:\n",
" print(f\"The elmenents in {compound} could not be added to the final list because they are not all relevant to the model.\")\n",
" else: \n",
" cur_values = [token.text for token in chunk if (token.ent_type_) or (token.pos_ == \"ADJ\")]\n",
" if (all(elem in lime_options for elem in cur_values) and (options is True)) or ((options is False)):\n",
" pos_options.extend(cur_values)\n",
" print(f\"From {chunk.text}, {cur_values} added to pos_options due to entity recognition.\") #for QA\n",
" elif len(chunk) >= 1:\n",
" cur_values = [token.text for token in chunk if token.pos_ in [\"NOUN\",\"ADJ\"]]\n",
" if (all(elem in lime_options for elem in cur_values) and (options is True)) or ((options is False)):\n",
" pos_options.extend(cur_values)\n",
" print(f\"From {chunk.text}, {cur_values} added to pos_options due to wildcard.\") #for QA\n",
" else:\n",
" print(f\"No options added for \\'{chunk.text}\\' \")\n",
" # Here I am going to try to pick up pronouns, which are people, and Adjectival Compliments.\n",
" for token in document:\n",
" if (token.text not in pos_options) and ((token.text in lime_options) or (options == False)):\n",
" #print(f\"executed {token.text} with {token.pos_} and {token.dep_}\") #QA\n",
" if (token.pos_ == \"ADJ\") and (token.dep_ in [\"acomp\",\"conj\"]):\n",
" pos_options.append(token.text) \n",
" elif (token.pos_ == \"PRON\") and (token.morph.get(\"PronType\")[0] == \"Prs\"):\n",
" pos_options.append(token.text)\n",
" \n",
" #Return the correct set of options based on user input, defaults to POS for simplicity.\n",
" if options:\n",
" return pos_options, lime_results\n",
" else:\n",
" return pos_options"
]
},
{
"cell_type": "code",
"execution_count": 20,
"id": "b04e7783-e51b-49b0-8165-afe1d5a1c576",
"metadata": {},
"outputs": [],
"source": [
"#Testing new code\n",
"a = \"People are fat and lazy.\"\n",
"b = \"I think she is beautiful.\"\n",
"doca = nlp(a)\n",
"docb = nlp(b)"
]
},
{
"cell_type": "code",
"execution_count": 21,
"id": "0a6bc521-9282-41ad-82c9-29e447d77635",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"No options added for 'People' \n"
]
},
{
"data": {
"text/plain": [
"['fat', 'lazy']"
]
},
"execution_count": 21,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"optsa, limea = critical_words(doca, True)\n",
"optsa"
]
},
{
"cell_type": "code",
"execution_count": 22,
"id": "042e94d3-65a5-4a20-b69a-96ec3296d7d4",
"metadata": {},
"outputs": [],
"source": [
"def lime_viz(df):\n",
" single_nearest = alt.selection_single(on='mouseover', nearest=True)\n",
" viz = alt.Chart(df).encode(\n",
" alt.X('Weight:Q', scale=alt.Scale(domain=(-1, 1))),\n",
" alt.Y('Word:N', sort='x', axis=None),\n",
" color=alt.Color(\"Weight\", scale=alt.Scale(scheme='blueorange', domain=[0], type=\"threshold\", range='diverging'), legend=None),\n",
" tooltip = (\"Word\",\"Weight\")\n",
" ).mark_bar().properties(title =\"Importance of individual words\")\n",
"\n",
" text = viz.mark_text(\n",
" fill=\"black\",\n",
" align='right',\n",
" baseline='middle'\n",
" ).encode(\n",
" text='Word:N'\n",
" )\n",
" limeplot = alt.LayerChart(layer=[viz,text], width = 300).configure_axis(grid=False).configure_view(strokeWidth=0)\n",
" return limeplot"
]
},
{
"cell_type": "code",
"execution_count": 23,
"id": "924eeea8-1d5d-4fe7-8308-164521919269",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"No options added for 'I' \n",
"From a white woman, ['white', 'woman'] added to pos_options due to wildcard.\n",
"From the street, ['street'] added to pos_options due to wildcard.\n",
"From an asian man, ['asian', 'man'] added to pos_options due to wildcard.\n"
]
},
{
"data": {
"text/plain": [
"['white', 'woman', 'street', 'asian', 'man', 'I']"
]
},
"execution_count": 23,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"test8 = \"I saw a white woman walking down the street with an asian man.\"\n",
"opts8, lime8 = critical_words(test8,True)\n",
"opts8"
]
},
{
"cell_type": "code",
"execution_count": 24,
"id": "734366df-ad99-4d80-87e1-51793e150681",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"<div id=\"altair-viz-adaa380d0d924bb594dd3aaee854acfd\"></div>\n",
"<script type=\"text/javascript\">\n",
" var VEGA_DEBUG = (typeof VEGA_DEBUG == \"undefined\") ? {} : VEGA_DEBUG;\n",
" (function(spec, embedOpt){\n",
" let outputDiv = document.currentScript.previousElementSibling;\n",
" if (outputDiv.id !== \"altair-viz-adaa380d0d924bb594dd3aaee854acfd\") {\n",
" outputDiv = document.getElementById(\"altair-viz-adaa380d0d924bb594dd3aaee854acfd\");\n",
" }\n",
" const paths = {\n",
" \"vega\": \"https://cdn.jsdelivr.net/npm//vega@5?noext\",\n",
" \"vega-lib\": \"https://cdn.jsdelivr.net/npm//vega-lib?noext\",\n",
" \"vega-lite\": \"https://cdn.jsdelivr.net/npm//vega-lite@4.17.0?noext\",\n",
" \"vega-embed\": \"https://cdn.jsdelivr.net/npm//vega-embed@6?noext\",\n",
" };\n",
"\n",
" function maybeLoadScript(lib, version) {\n",
" var key = `${lib.replace(\"-\", \"\")}_version`;\n",
" return (VEGA_DEBUG[key] == version) ?\n",
" Promise.resolve(paths[lib]) :\n",
" new Promise(function(resolve, reject) {\n",
" var s = document.createElement('script');\n",
" document.getElementsByTagName(\"head\")[0].appendChild(s);\n",
" s.async = true;\n",
" s.onload = () => {\n",
" VEGA_DEBUG[key] = version;\n",
" return resolve(paths[lib]);\n",
" };\n",
" s.onerror = () => reject(`Error loading script: ${paths[lib]}`);\n",
" s.src = paths[lib];\n",
" });\n",
" }\n",
"\n",
" function showError(err) {\n",
" outputDiv.innerHTML = `<div class=\"error\" style=\"color:red;\">${err}</div>`;\n",
" throw err;\n",
" }\n",
"\n",
" function displayChart(vegaEmbed) {\n",
" vegaEmbed(outputDiv, spec, embedOpt)\n",
" .catch(err => showError(`Javascript Error: ${err.message}<br>This usually means there's a typo in your chart specification. See the javascript console for the full traceback.`));\n",
" }\n",
"\n",
" if(typeof define === \"function\" && define.amd) {\n",
" requirejs.config({paths});\n",
" require([\"vega-embed\"], displayChart, err => showError(`Error loading script: ${err.message}`));\n",
" } else {\n",
" maybeLoadScript(\"vega\", \"5\")\n",
" .then(() => maybeLoadScript(\"vega-lite\", \"4.17.0\"))\n",
" .then(() => maybeLoadScript(\"vega-embed\", \"6\"))\n",
" .catch(showError)\n",
" .then(() => displayChart(vegaEmbed));\n",
" }\n",
" })({\"config\": {\"view\": {\"continuousWidth\": 400, \"continuousHeight\": 300, \"strokeWidth\": 0}, \"axis\": {\"grid\": false}}, \"layer\": [{\"mark\": \"bar\", \"encoding\": {\"color\": {\"field\": \"Weight\", \"legend\": null, \"scale\": {\"domain\": [0], \"range\": \"diverging\", \"scheme\": \"blueorange\", \"type\": \"threshold\"}, \"type\": \"quantitative\"}, \"tooltip\": [{\"field\": \"Word\", \"type\": \"nominal\"}, {\"field\": \"Weight\", \"type\": \"quantitative\"}], \"x\": {\"field\": \"Weight\", \"scale\": {\"domain\": [-1, 1]}, \"type\": \"quantitative\"}, \"y\": {\"axis\": null, \"field\": \"Word\", \"sort\": \"x\", \"type\": \"nominal\"}}, \"title\": \"Importance of individual words\"}, {\"mark\": {\"type\": \"text\", \"align\": \"right\", \"baseline\": \"middle\", \"fill\": \"black\"}, \"encoding\": {\"color\": {\"field\": \"Weight\", \"legend\": null, \"scale\": {\"domain\": [0], \"range\": \"diverging\", \"scheme\": \"blueorange\", \"type\": \"threshold\"}, \"type\": \"quantitative\"}, \"text\": {\"field\": \"Word\", \"type\": \"nominal\"}, \"tooltip\": [{\"field\": \"Word\", \"type\": \"nominal\"}, {\"field\": \"Weight\", \"type\": \"quantitative\"}], \"x\": {\"field\": \"Weight\", \"scale\": {\"domain\": [-1, 1]}, \"type\": \"quantitative\"}, \"y\": {\"axis\": null, \"field\": \"Word\", \"sort\": \"x\", \"type\": \"nominal\"}}, \"title\": \"Importance of individual words\"}], \"data\": {\"name\": \"data-d686d7fc533c26b0bdc6066e4351f840\"}, \"width\": 300, \"$schema\": \"https://vega.github.io/schema/vega-lite/v4.17.0.json\", \"datasets\": {\"data-d686d7fc533c26b0bdc6066e4351f840\": [{\"Word\": \"with\", \"Weight\": 0.3289028288853927}, {\"Word\": \"woman\", \"Weight\": -0.26094440033196564}, {\"Word\": \"asian\", \"Weight\": 0.24561077002890458}, {\"Word\": \"walking\", \"Weight\": 0.19194218998931795}, {\"Word\": \"white\", \"Weight\": -0.14942503537339621}, {\"Word\": \"down\", \"Weight\": -0.14547403123420313}, {\"Word\": \"the\", \"Weight\": 0.14096934306553166}, {\"Word\": \"I\", \"Weight\": -0.08672932329874143}, {\"Word\": \"street\", \"Weight\": 0.06704680513000527}, {\"Word\": \"a\", \"Weight\": -0.03171807940472653}, {\"Word\": \"an\", \"Weight\": -0.006746730007490843}, {\"Word\": \"saw\", \"Weight\": 0.0019276122088497296}, {\"Word\": \"man\", \"Weight\": -0.0005652423244728638}]}}, {\"mode\": \"vega-lite\"});\n",
"</script>"
],
"text/plain": [
"alt.LayerChart(...)"
]
},
"execution_count": 24,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"lime_viz(lime8)"
]
},
{
"cell_type": "code",
"execution_count": 25,
"id": "816e1c4b-7f02-41b1-b430-2f3750ae6c4a",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"No options added for 'I' \n",
"From a white woman, ['white', 'woman'] added to pos_options due to wildcard.\n",
"From the street, ['street'] added to pos_options due to wildcard.\n",
"From an asian man, ['asian', 'man'] added to pos_options due to wildcard.\n"
]
}
],
"source": [
"probability, sentiment = eval_pred_test(test8, return_all=True)\n",
"options, lime = critical_words(test8,options=True)"
]
},
{
"cell_type": "code",
"execution_count": 38,
"id": "a437a4eb-73b3-4b3c-a719-8dde2ad6dd3c",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"From I, [] added to pos_options due to wildcard.\n",
"From men, ['men'] added to pos_options due to wildcard.\n",
"From women, ['women'] added to pos_options due to wildcard.\n",
"From the same respect, ['same', 'respect'] added to pos_options due to wildcard.\n"
]
}
],
"source": [
"bug = \"I find men and women deserve the same respect.\"\n",
"options = critical_words(bug)"
]
},
{
"cell_type": "code",
"execution_count": 29,
"id": "8676defd-0908-4218-a1d6-218de3fb7119",
"metadata": {},
"outputs": [],
"source": [
"bug_doc = nlp(bug)"
]
},
{
"cell_type": "code",
"execution_count": 35,
"id": "21b9e39b-2fcd-4c6f-8fe6-0d571cd79cca",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"I\n",
"PRON\n",
"a man\n",
"NOUN\n",
"woman\n",
"NOUN\n",
"the same respect\n",
"NOUN\n"
]
}
],
"source": [
"for chunk in bug_doc.noun_chunks:\n",
" print(chunk.text)\n",
" print(chunk[-1].pos_)"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "38279d2d-e763-4329-a65e-1a67d6f5ebb8",
"metadata": {},
"outputs": [],
"source": []
}
],
"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": "3.8.8"
}
},
"nbformat": 4,
"nbformat_minor": 5
}
|