{ "cells": [ { "cell_type": "code", "execution_count": 1, "id": "169e11e8-19c7-47ae-b8af-817e2e474ee7", "metadata": { "tags": [] }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Overwriting book.jsgf\n" ] } ], "source": [ "%%writefile book.jsgf\n", "#JSGF V1.0 UTF-8 pl;\n", "\n", "grammar book;\n", "\n", "public = chcialbym zarezerwowac stolik ;\n", "\n", " = na {day};\n", "\n", " = dzisiaj | jutro | poniedzialek | wtorek | srode | czwartek | piatek | sobote | niedziele;\n", "\n", " = na [godzine] {hour};\n", "\n", " = [];\n", "\n", " = dziewiata | dziesiata | jedenasta | dwunasta;\n", "\n", " = pietnascie | trzydziesci;\n", "\n", " = (na | dla) {size} osob;\n", "\n", " = dwie | dwoch | trzy | trzech | cztery | czterech | piec | pieciu;" ] }, { "cell_type": "code", "execution_count": 2, "id": "88cc9879-a59c-4c90-adb6-1022c42913e6", "metadata": { "tags": [] }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Overwriting order.jsgf\n" ] } ], "source": [ "%%writefile order.jsgf\n", "#JSGF V1.0 UTF-8 pl;\n", "\n", "grammar order;\n", "\n", "public = (chcialbym | chcialabym) zamowic ;\n", "\n", " = {dish};\n", "\n", " = pizza peperoni | spaghetti carbonara | burger szefa;\n", "\n", " = (do stolika | na stolik ) {stolik};\n", "\n", " = 100 | 101 | 102 | 103 | 104 | 105;" ] }, { "cell_type": "code", "execution_count": 3, "id": "c5ffadfe-ea18-45ab-a9f9-b2cfd3c2fee7", "metadata": { "tags": [] }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Overwriting recommend.jsgf\n" ] } ], "source": [ "%%writefile recommend.jsgf\n", "#JSGF V1.0 UTF-8 pl;\n", "\n", "grammar recommend;\n", "\n", "public = (jakie|jaki|jaka|jakiego) polecasz;\n", "\n", " = {food_or_drink};\n", "\n", " = jedzenia | picia | napoj | danie | dania;\n", "\n", "public = ((jakie|jaki|jaka|jakiego) ) *;\n", "\n", "public = ((jakie|jaki|jaka|jakiego) ) *;\n", "\n", " = {dish};\n", "\n", " = makarony | makaron | pizze | pizza | burgery | burger | salatki | salatka | deser | desery;\n", "\n", " = {drink};\n", "\n", " = piwo | piwa | wino | wina | sok | soki | smoothie | kawe | kawy | kawe | herbate | herbata | herbaty;" ] }, { "cell_type": "code", "execution_count": 4, "id": "57fb9886-0232-4dd9-b385-7cae1b068f6e", "metadata": { "tags": [] }, "outputs": [ { "data": { "text/plain": [ "Grammar(version=1.0, charset=UTF-8, language=pl, name=order)" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "import jsgf\n", "\n", "book_grammar = jsgf.parse_grammar_file('book.jsgf')\n", "order_grammar = jsgf.parse_grammar_file('order.jsgf')\n", "recommend_grammar = jsgf.parse_grammar_file('recommend.jsgf')\n", "order_grammar" ] }, { "cell_type": "code", "execution_count": 5, "id": "8f5acde6-4f3c-425d-a202-f361b219922d", "metadata": { "tags": [] }, "outputs": [ { "data": { "text/plain": [ "[Rule(name='rezerwuj', visible=True, expansion=Sequence(Literal('chcialbym zarezerwowac stolik'), NamedRuleRef('dzien_rezerwacji'), NamedRuleRef('godzina_rezerwacji'), NamedRuleRef('liczba_osob')))]" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "utterance = 'chcialbym zarezerwowac stolik na jutro na godzine dwunasta trzydziesci na piec osob'\n", "matched = book_grammar.find_matching_rules(utterance)\n", "if not matched:\n", " matched = order_grammar.find_matching_rules(utterance)\n", "if not matched:\n", " matched = recommend_grammar.find_matching_rules(utterance)\n", "matched" ] }, { "cell_type": "code", "execution_count": 6, "id": "6cb7d2ba-d8ee-403a-bbf3-d7ea3e0475ba", "metadata": { "tags": [] }, "outputs": [ { "data": { "text/plain": [ "[Rule(name='zamow', visible=True, expansion=Sequence(ParsedAlternativeSet(Literal('chcialbym'), Literal('chcialabym')), Literal('zamowic'), NamedRuleRef('danie'), NamedRuleRef('stolik')))]" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "utterance = 'chcialbym zamowic burger szefa do stolika 104'\n", "matched = book_grammar.find_matching_rules(utterance)\n", "if not matched:\n", " matched = order_grammar.find_matching_rules(utterance)\n", "if not matched:\n", " matched = recommend_grammar.find_matching_rules(utterance)\n", "matched" ] }, { "cell_type": "code", "execution_count": 7, "id": "371ee857-418d-4546-8e29-88b6c8706659", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[Rule(name='polec_danie', visible=True, expansion=KleeneStar(RequiredGrouping(Sequence(RequiredGrouping(ParsedAlternativeSet(Literal('jakie'), Literal('jaki'), Literal('jaka'), Literal('jakiego'))), NamedRuleRef('danie'))))),\n", " Rule(name='polec_napoj', visible=True, expansion=KleeneStar(RequiredGrouping(Sequence(RequiredGrouping(ParsedAlternativeSet(Literal('jakie'), Literal('jaki'), Literal('jaka'), Literal('jakiego'))), NamedRuleRef('napoj')))))]" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ "utterance = 'jakie makarony polecasz'\n", "matched = book_grammar.find_matching_rules(utterance)\n", "if not matched:\n", " matched = order_grammar.find_matching_rules(utterance)\n", "if not matched:\n", " matched = recommend_grammar.find_matching_rules(utterance)\n", "matched" ] }, { "cell_type": "code", "execution_count": 8, "id": "c914cc3d-b297-4e05-917c-6a2f09c0f8a9", "metadata": { "tags": [] }, "outputs": [ { "data": { "text/plain": [ "{'act': 'recommend', 'slots': [('dish', 'makarony')]}" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "def get_dialog_act(rule):\n", " slots = []\n", " get_slots(rule.expansion, slots)\n", " return {'act': rule.grammar.name, 'slots': slots}\n", "\n", "def get_slots(expansion, slots):\n", " if expansion.tag != '':\n", " slots.append((expansion.tag, expansion.current_match))\n", " return\n", "\n", " for child in expansion.children:\n", " get_slots(child, slots)\n", "\n", " if not expansion.children and isinstance(expansion, jsgf.NamedRuleRef):\n", " get_slots(expansion.referenced_rule.expansion, slots)\n", "\n", "get_dialog_act(matched[0])" ] }, { "cell_type": "code", "execution_count": 13, "id": "91eaed10-e8c3-4429-a232-d317b84ec38f", "metadata": { "tags": [] }, "outputs": [ { "data": { "text/plain": [ "{'act': 'book',\n", " 'slots': [('day', 'jutro'), ('hour', 'dziesiata'), ('size', 'trzech')]}" ] }, "execution_count": 13, "metadata": {}, "output_type": "execute_result" } ], "source": [ "def nlu(utterance):\n", " matched = book_grammar.find_matching_rules(utterance)\n", " if not matched:\n", " matched = order_grammar.find_matching_rules(utterance)\n", " if not matched:\n", " matched = recommend_grammar.find_matching_rules(utterance)\n", "\n", " if matched:\n", " return get_dialog_act(matched[0])\n", " else:\n", " return {'act': 'null', 'slots': []}\n", "\n", "nlu('chcialbym zarezerwowac stolik na jutro na godzine dziesiata dla trzech osob')" ] }, { "cell_type": "code", "execution_count": 14, "id": "d85e8543-be59-4c5f-b9f9-27631a899c2a", "metadata": { "tags": [] }, "outputs": [ { "data": { "text/plain": [ "{'act': 'recommend', 'slots': [('dish', 'makarony')]}" ] }, "execution_count": 14, "metadata": {}, "output_type": "execute_result" } ], "source": [ "nlu('jakie makarony polecasz')" ] } ], "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": "3.10.0" } }, "nbformat": 4, "nbformat_minor": 5 }