Copy_of_Week_03_Agents.ipynb

{"cells":[{"cell_type":"markdown","metadata":{"id":"reej2tDWSyeX"},"source":["Vaccum Cleaning Agent with locations A and B"]},{"cell_type":"markdown","metadata":{"id":"OmRycx4GSyeY"},"source":["![vc.png](attachment:vc.png)"]},{"cell_type":"markdown","source":["# Some important things:\n","\n","1. Agent Description: The aim of the agent .....\n","2. List the sensors: Location, Status\n","3. Sensors' values:\n",">Location -> {A, B}, Status -> {Dirty, Clean}\n","4. List the actions: Suck, Right, Left\n","5. Agent function: **perception --> action**\n","> [A, Clean] --> Right <br>\n","> [A, Dirty] --> Suck<br>\n","> [B, Clean] --> Left<br>\n","> [B, Dirty] --> Suck<br>\n","\n"],"metadata":{"id":"JB_wyml9WUg4"}},{"cell_type":"markdown","metadata":{"id":"jKEeqHbzSyeQ"},"source":["# Table driven agent"]},{"cell_type":"code","source":["table={('A','Clean'):'Right',\n","       ('A','Dirty'):'Suck',\n","       ('B','Clean'):'Left',\n","       ('B','Dirty'):'Suck'}"],"metadata":{"id":"sdVAkIfVfPLl"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["percepts=[]  # to store percept sequence\n","def table_driven_agent(percept):\n","    print('Perception Received: '+ str(percept))\n","    percepts.append(percept) # updating percept history\n","    action = lookup(percept,table) # searching for action\n","    return action\n","\n","def lookup(p,t):\n","    return t[p]"],"metadata":{"id":"-66u3QD8fbIj"},"execution_count":null,"outputs":[]},{"cell_type":"code","execution_count":null,"metadata":{"id":"oYvL-4cuSyeZ","outputId":"5a82b7b4-a442-4083-8108-f43c154cff20","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1655461435476,"user_tz":-360,"elapsed":33374,"user":{"displayName":"Raihan Ahmed","userId":"17078036816298851284"}}},"outputs":[{"name":"stdout","output_type":"stream","text":["Perception Received: ('B', 'Dirty')\n","Action Performed: Suck\n","Get Perception (yes/no): yes\n","Perception Received: ('B', 'Clean')\n","Action Performed: Left\n","Get Perception (yes/no): yes\n","Perception Received: ('A', 'Clean')\n","Action Performed: Right\n","Get Perception (yes/no): no\n"]}],"source":["# lets simulate the agent\n","import random\n","Location = random.choice(['A','B'])\n","Condition = random.choice(['Clean','Dirty'])\n","\n","while True: # to perceieve environment repeatedly\n","    action = table_driven_agent((Location, Condition))\n","    print('Action Performed: '+ action)\n","    cmd = input('Get Perception (yes/no): ')\n","    if(cmd != 'yes'): break\n","    if action == 'Right':\n","        Location = 'B'\n","        Condition = random.choice(['Clean','Dirty'])\n","    elif action == 'Left':\n","        Location = 'A'\n","        Condition = random.choice(['Clean','Dirty'])\n","    else:\n","        Condition = 'Clean'"]},{"cell_type":"markdown","metadata":{"id":"yfOFRdsmSyec"},"source":["# Simple reflex agent"]},{"cell_type":"code","source":["def simple_reflex_agent(percept):\n","    print('Perception Received: '+ str(percept))\n","    location = percept[0]\n","    status = percept[1]\n","    if status =='Dirty':\n","        action = 'Suck'\n","    elif location == 'A':\n","        action = 'Right'\n","    elif location =='B':\n","        action = 'Left'\n","    return action"],"metadata":{"id":"WRQ3RsmchDht"},"execution_count":null,"outputs":[]},{"cell_type":"code","execution_count":null,"metadata":{"id":"GfEzFJTVSyed","outputId":"bc4e8ece-ddb3-46f6-eb18-15c2a9d7f23b","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1655461477163,"user_tz":-360,"elapsed":41731,"user":{"displayName":"Raihan Ahmed","userId":"17078036816298851284"}}},"outputs":[{"name":"stdout","output_type":"stream","text":["Perception Received: ('B', 'Clean')\n","Action Performed: Left\n","Get Perception (yes/no): yes\n","Perception Received: ('A', 'Dirty')\n","Action Performed: Suck\n","Get Perception (yes/no): yes\n","Perception Received: ('A', 'Clean')\n","Action Performed: Right\n","Get Perception (yes/no): no\n"]}],"source":["import random\n","Location = random.choice(['A','B'])\n","Condition = random.choice(['Clean','Dirty'])\n","\n","while True:\n","    action= simple_reflex_agent((Location,Condition))\n","    print('Action Performed: '+ action)\n","    cmd = input('Get Perception (yes/no): ')\n","    if(cmd == 'no' or cmd != 'yes'): break\n","    if action == 'Right':\n","        Location = 'B'\n","        Condition = random.choice(['Clean','Dirty'])\n","    elif action== 'Left':\n","        Location = 'A'\n","        Condition = random.choice(['Clean','Dirty'])\n","    else:\n","        Condition = 'Clean'"]},{"cell_type":"markdown","metadata":{"id":"wA7yF32tSyee"},"source":["# Model based agent"]},{"cell_type":"code","source":["model = {'A':'Unknown', 'B':'Unknown'}\n","world_state = 'Bad'\n","action = 'Unknown'\n","\n","def update_state(action, percept, model):\n","    location = percept[0]\n","    status = percept[1]\n","    model[location] = status #udate the model based on current perception\n","\n","    global world_state\n","    #udate the model based on action\n","    if action == 'Suck':\n","        model[location]='Clean'\n","    # model checking to update world state\n","    if model['A']=='Clean' and model['B']=='Clean':\n","        world_state = 'Good'\n","    else:\n","        world_state = 'Bad'\n","    return world_state\n","\n","\n","def model_based_reflex_agent(percept):\n","    location = percept[0]\n","    status = percept[1]\n","\n","    global world_state,action,model\n","\n","    if world_state == 'Good':\n","        action = 'Pause'\n","        return action\n","    elif status == 'Dirty':\n","        action = 'Suck'\n","    elif location == 'A':\n","        action = 'Right'\n","    elif location == 'B':\n","        action = 'Left'\n","    world_state = update_state(action, percept, model)\n","    print('Perception: '+str(percept))\n","    print('Action Performed: '+ action)\n","    print('Model: '+str(model))\n","    print('State: '+str(world_state))\n","    return action"],"metadata":{"id":"KGvoakXKiaqT"},"execution_count":null,"outputs":[]},{"cell_type":"code","execution_count":null,"metadata":{"id":"djwVDNKgSyef","outputId":"8a13a01e-728d-4602-947c-e435b9a7d210","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1655461486647,"user_tz":-360,"elapsed":9545,"user":{"displayName":"Raihan Ahmed","userId":"17078036816298851284"}}},"outputs":[{"name":"stdout","output_type":"stream","text":["*****\n","Perception: ('A', 'Clean')\n","Action Performed: Right\n","Model: {'A': 'Clean', 'B': 'Unknown'}\n","State: Bad\n","*****\n","Perception: ('B', 'Dirty')\n","Action Performed: Suck\n","Model: {'A': 'Clean', 'B': 'Clean'}\n","State: Good\n","*****\n","Stopped. Do restart? (yes/no): no\n"]}],"source":["import random\n","Location = random.choice(['A','B'])\n","Condition= random.choice(['Clean','Dirty'])\n","\n","while True:\n","    print('*****')\n","    action = model_based_reflex_agent((Location,Condition))\n","    if action == 'Right':\n","        Location = 'B'\n","        Condition = random.choice(['Clean','Dirty'])\n","    elif action == 'Left':\n","        Location = 'A'\n","        Condition = random.choice(['Clean','Dirty'])\n","    elif action == 'Suck':\n","        Condition = 'Clean'\n","    elif action == 'Pause':\n","        cmd = input('Stopped. Do restart? (yes/no): ')\n","        if(cmd == 'no' or cmd != 'yes'): break\n","        Location = random.choice(['A','B'])\n","        Condition = random.choice(['Clean','Dirty'])\n","        model = {'A':'Unknown', 'B':'Unknown'}\n","        world_state = 'Bad'\n","        action = 'Unknown'"]},{"cell_type":"markdown","metadata":{"id":"SduRNewjSyeg"},"source":["# **Exercise 1**"]},{"cell_type":"markdown","metadata":{"id":"XQfOSnycSyeh"},"source":["Implement diffierent types of agent implementation to do a certain task of your choice.\n","1. Table Driven Agent\n","2. Simple Reflex Agent\n","\n","<p><b>Before implementation describe the following:</b></p>\n","\n","1. Agent Description: The aim of the agent .....\n","2. List the sensors:\n","3. Sensors' values:\n","3. List the actions:\n","4. Agent function: perception --> action"]},{"cell_type":"markdown","metadata":{"id":"9s44XzAtSyei"},"source":[]},{"cell_type":"code","execution_count":null,"metadata":{"id":"uKEfWWEgSyei"},"outputs":[],"source":[]}],"metadata":{"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"codemirror_mode":{"name":"ipython","version":3},"file_extension":".py","mimetype":"text/x-python","name":"python","nbconvert_exporter":"python","pygments_lexer":"ipython3","version":"3.8.8"},"colab":{"provenance":[{"file_id":"1wwikMVsuTMac2hsXpVA12e82PTeK651J","timestamp":1655462534538}]}},"nbformat":4,"nbformat_minor":0}