Generic_DialogSystem/DialogManager.ipynb
2023-06-02 12:43:15 +02:00

296 lines
9.2 KiB
Plaintext

{
"cells": [
{
"cell_type": "code",
"execution_count": 48,
"metadata": {},
"outputs": [],
"source": [
"import json\n",
"import os\n",
"\n",
"\n",
"class DST():\n",
" def __init__(self):\n",
" self.state = json.load(open('dictionary.json'))\n",
"\n",
" def update(self, user_act=None):\n",
" for intent, domain, slot, value in user_act:\n",
" domain = domain.lower()\n",
" intent = intent.lower()\n",
" slot = slot.lower()\n",
" \n",
" if domain not in self.state['belief_state']:\n",
" continue\n",
"\n",
" if intent == 'inform':\n",
" if slot == 'none' or slot == '':\n",
" continue\n",
"\n",
" domain_dic = self.state['belief_state'][domain]\n",
"\n",
" if slot in domain_dic:\n",
" self.state['belief_state'][domain][slot] = value\n",
"\n",
" elif intent == 'request':\n",
" if domain not in self.state['request_state']:\n",
" self.state['request_state'][domain] = {}\n",
" if slot not in self.state['request_state'][domain]:\n",
" self.state['request_state'][domain][slot] = 0\n",
" \n",
" self.state['user_act'] = user_act\n",
" return self.state\n",
" def init_session(self):\n",
" self.state = json.load(open('dictionary.json'))"
]
},
{
"cell_type": "code",
"execution_count": 49,
"metadata": {},
"outputs": [],
"source": [
"dst = DST()\n",
"user_act = [('inform', 'payment', 'type', 'karta'), ('inform', 'delivery', 'type','paczkomat'), ('inform', 'product', 'type', 'telefon'), ('request', 'product', 'type', '?')]\n",
"state = dst.update(user_act)"
]
},
{
"cell_type": "code",
"execution_count": 50,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"{'payment': {'type': 'karta', 'amount': '', 'loyalty_card': ''}, 'delivery': {'type': 'paczkomat', 'address': '', 'time': ''}, 'product': {'name': '', 'type': 'telefon', 'brand': '', 'price_range': '', 'price': '', 'quantity': '', 'quality': ''}}\n",
"{'product': {'type': 0}}\n"
]
},
{
"data": {
"text/plain": [
"{'user_act': [('inform', 'payment', 'type', 'karta'),\n",
" ('inform', 'delivery', 'type', 'paczkomat'),\n",
" ('inform', 'product', 'type', 'telefon'),\n",
" ('request', 'product', 'type', '?')],\n",
" 'system_act': [],\n",
" 'belief_state': {'payment': {'type': 'karta',\n",
" 'amount': '',\n",
" 'loyalty_card': ''},\n",
" 'delivery': {'type': 'paczkomat', 'address': '', 'time': ''},\n",
" 'product': {'name': '',\n",
" 'type': 'telefon',\n",
" 'brand': '',\n",
" 'price_range': '',\n",
" 'price': '',\n",
" 'quantity': '',\n",
" 'quality': ''}},\n",
" 'request_state': {'product': {'type': 0}},\n",
" 'terminated': False,\n",
" 'history': []}"
]
},
"execution_count": 50,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"print(state['belief_state'])\n",
"print(state['request_state'])\n",
"dst.state"
]
},
{
"cell_type": "code",
"execution_count": 51,
"metadata": {},
"outputs": [],
"source": [
"from collections import defaultdict\n",
"from convlab.policy.policy import Policy\n",
"import json\n",
"\n",
"class SimpleRulePolicy(Policy):\n",
" def __init__(self):\n",
" Policy.__init__(self)\n",
" self.db = json.load(open('product_db.json'))\n",
"\n",
" def predict(self, state):\n",
" self.results = []\n",
" system_action = defaultdict(list)\n",
" user_action = defaultdict(list)\n",
" system_acts = []\n",
" for intent, domain, slot, value in state['user_act']:\n",
" user_action[(domain.lower(), intent.lower())].append((slot.lower(), value))\n",
" for user_act in user_action:\n",
" self.update_system_action(user_act, user_action, state, system_action)\n",
" system_acts = [[intent, domain, slot, value] for (domain, intent), slots in system_action.items() for slot, value in slots]\n",
" state['system_act'] = system_acts\n",
" return system_acts\n",
"\n",
"\n",
" def update_system_action(self, user_act, user_action, state, system_action):\n",
" domain, intent = user_act\n",
" constraints = [(slot, value) for slot, value in state['belief_state'][domain.lower()].items() if value != '']\n",
" self.results = self.db['database'][domain]\n",
"\n",
" # Reguła 1\n",
" if intent == 'request':\n",
" if len(self.results) == 0:\n",
" system_action[(domain, 'NoOffer')] = []\n",
" else:\n",
" for slot in user_action[user_act]:\n",
" if self.results and slot[0] in self.results[0]:\n",
" system_action[(domain, 'Inform')].append([slot[0], self.results[0].get(slot[0], 'unknown')])\n",
"\n",
" # Reguła 2\n",
" elif intent == 'inform':\n",
" if len(self.results) == 0:\n",
" system_action[(domain, 'NoOffer')] = []\n",
" else:\n",
" system_action[(domain, 'Inform')].append(['Choice', str(len(self.results))])\n",
" for product in self.results:\n",
" if all(product.get(slot, '').lower() == value.lower() for slot, value in constraints):\n",
" system_action[(domain, 'Recommend')].append(['Name', product['name']])\n",
" break\n",
"\n"
]
},
{
"cell_type": "code",
"execution_count": 52,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"[['Inform', 'product', 'Choice', '11'],\n",
" ['Recommend', 'product', 'Name', 'RedBull']]"
]
},
"execution_count": 52,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"dst = DST()\n",
"user_act = [('inform', 'product', 'type', 'energol')]\n",
"state = dst.update(user_act)\n",
"policy = SimpleRulePolicy()\n",
"policy.predict(state)"
]
},
{
"cell_type": "code",
"execution_count": 53,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"WARNING:root:nlu info_dict is not initialized\n",
"WARNING:root:dst info_dict is not initialized\n",
"WARNING:root:policy info_dict is not initialized\n",
"WARNING:root:nlg info_dict is not initialized\n"
]
}
],
"source": [
"from convlab.dialog_agent import PipelineAgent\n",
"policy = SimpleRulePolicy()\n",
"agent = PipelineAgent(nlu=None, dst=dst, policy=policy, nlg=None, name='sys')"
]
},
{
"cell_type": "code",
"execution_count": 54,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"[['Inform', 'product', 'Choice', '11'],\n",
" ['Recommend', 'product', 'Name', 'pomidor']]"
]
},
"execution_count": 54,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"agent.response([('inform', 'product', 'type', 'warzywo'), ('inform', 'product', 'price_range', 'tani'), ('inform', 'product', 'quality', 'exquisite')])"
]
},
{
"cell_type": "code",
"execution_count": 55,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"[['Inform', 'product', 'Choice', '11'],\n",
" ['Recommend', 'product', 'Name', 'Sok pomarańczowy']]"
]
},
"execution_count": 55,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"agent.response([('inform', 'product', 'type', 'napój'), ('inform', 'product', 'price_range', 'drogi'), ('inform', 'product', 'quality', 'exquisite')])"
]
},
{
"cell_type": "code",
"execution_count": 56,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"[['Inform', 'product', 'Choice', '11'],\n",
" ['Recommend', 'product', 'Name', 'banan']]"
]
},
"execution_count": 56,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"agent.response([('inform', 'product', 'type', 'owoc'), ('inform', 'product', 'price_range', 'tani'), ('inform', 'product', 'quality', 'exquisite')])"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "py38",
"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": "3.8.16"
},
"orig_nbformat": 4
},
"nbformat": 4,
"nbformat_minor": 2
}