anon_env.py

'''
Interactions with CityFlow, get/set values from CityFlow, pass it to RL agents
'''

import pickle
import numpy as np
import json
import sys
import pandas as pd
import os
import cityflow as engine
import time
import threading
from multiprocessing import Process, Pool
from script import get_traffic_volume
from copy import deepcopy

class RoadNet:

    def __init__(self, roadnet_file):
        self.roadnet_dict = json.load(open(roadnet_file,"r"))
        self.net_edge_dict = {}
        self.net_node_dict = {}
        self.net_lane_dict = {}

        self.generate_node_dict()
        self.generate_edge_dict()
        self.generate_lane_dict()

    def generate_node_dict(self):
        '''
        node dict has key as node id, value could be the dict of input nodes and output nodes
        :return:
        '''

        for node_dict in self.roadnet_dict['intersections']:
            node_id = node_dict['id']
            road_links = node_dict['roads']
            input_nodes = []
            output_nodes = []
            input_edges = []
            output_edges = {}
            for road_link_id in road_links:
                road_link_dict = self._get_road_dict(road_link_id)
                if road_link_dict['startIntersection'] == node_id:
                    end_node = road_link_dict['endIntersection']
                    output_nodes.append(end_node)
                    # todo add output edges
                elif road_link_dict['endIntersection'] == node_id:
                    input_edges.append(road_link_id)
                    start_node = road_link_dict['startIntersection']
                    input_nodes.append(start_node)
                    output_edges[road_link_id] = set()
                    pass

            # update roadlinks
            actual_roadlinks = node_dict['roadLinks']
            for actual_roadlink in actual_roadlinks:
                output_edges[actual_roadlink['startRoad']].add(actual_roadlink['endRoad'])

            net_node = {
                'node_id': node_id,
                'input_nodes': list(set(input_nodes)),
                'input_edges': list(set(input_edges)),
                'output_nodes': list(set(output_nodes)),
                'output_edges': output_edges# should be a dict, with key as an input edge, value as output edges
            }
            if node_id not in self.net_node_dict.keys():
                self.net_node_dict[node_id] = net_node

    def _get_road_dict(self, road_id):
        for item in self.roadnet_dict['roads']:
            if item['id'] == road_id:
                return item
        print("Cannot find the road id {0}".format(road_id))
        sys.exit(-1)
        # return None

    def generate_edge_dict(self):
        '''
        edge dict has key as edge id, value could be the dict of input edges and output edges
        :return:
        '''
        for edge_dict in self.roadnet_dict['roads']:
            edge_id = edge_dict['id']
            input_node = edge_dict['startIntersection']
            output_node = edge_dict['endIntersection']

            net_edge = {
                'edge_id': edge_id,
                'input_node': input_node,
                'output_node': output_node,
                'input_edges': self.net_node_dict[input_node]['input_edges'],
                'output_edges': self.net_node_dict[output_node]['output_edges'][edge_id],

            }
            if edge_id not in self.net_edge_dict.keys():
                self.net_edge_dict[edge_id] = net_edge

    def generate_lane_dict(self):
        lane_dict = {}
        for node_dict in self.roadnet_dict['intersections']:
            for road_link in node_dict["roadLinks"]:
                lane_links = road_link["laneLinks"]
                start_road = road_link["startRoad"]
                end_road = road_link["endRoad"]
                for lane_link in lane_links:
                    start_lane = start_road + "_" + str(lane_link['startLaneIndex'])
                    end_lane = end_road + "_" +str(lane_link["endLaneIndex"])
                    if start_lane not in lane_dict:
                        lane_dict[start_lane] = {
                            "output_lanes": [end_lane],
                            "input_lanes": []
                        }
                    else:
                        lane_dict[start_lane]["output_lanes"].append(end_lane)
                    if end_lane not in lane_dict:
                        lane_dict[end_lane] = {
                            "output_lanes": [],
                            "input_lanes": [start_lane]
                        }
                    else:
                        lane_dict[end_lane]["input_lanes"].append(start_lane)

        self.net_lane_dict = lane_dict

    def hasEdge(self, edge_id):
        if edge_id in self.net_edge_dict.keys():
            return True
        else:
            return False

    def getEdge(self, edge_id):
        if edge_id in self.net_edge_dict.keys():
            return edge_id
        else:
            return None

    def getOutgoing(self, edge_id):
        if edge_id in self.net_edge_dict.keys():
            return self.net_edge_dict[edge_id]['output_edges']
        else:
            return []


class Intersection:
    DIC_PHASE_MAP = {
        0: 1,
        1: 2,
        2: 3,
        3: 4,
        -1: 0
    }
    def __init__(self, inter_id, dic_traffic_env_conf, eng, light_id_dict,path_to_log):
        self.inter_id = inter_id

        self.inter_name = "intersection_{0}_{1}".format(inter_id[0], inter_id[1])

        self.eng = eng

        self.fast_compute = dic_traffic_env_conf['FAST_COMPUTE']

        self.controlled_model  = dic_traffic_env_conf['MODEL_NAME']
        self.path_to_log = path_to_log

        # =====  intersection settings =====
        self.list_approachs = ["W", "E", "N", "S"]
        self.dic_approach_to_node = {"W": 2, "E": 0, "S": 3, "N": 1}
        # self.dic_entering_approach_to_edge = {
        #    approach: "road{0}_{1}_{2}".format(self.dic_approach_to_node[approach], light_id) for approach in self.list_approachs}

        self.dic_entering_approach_to_edge = {"W": "road_{0}_{1}_0".format(inter_id[0] - 1, inter_id[1])}
        self.dic_entering_approach_to_edge.update({"E": "road_{0}_{1}_2".format(inter_id[0] + 1, inter_id[1])})
        self.dic_entering_approach_to_edge.update({"S": "road_{0}_{1}_1".format(inter_id[0], inter_id[1] - 1)})
        self.dic_entering_approach_to_edge.update({"N": "road_{0}_{1}_3".format(inter_id[0], inter_id[1] + 1)})

        self.dic_exiting_approach_to_edge = {
            approach: "road_{0}_{1}_{2}".format(inter_id[0], inter_id[1], self.dic_approach_to_node[approach]) for
        approach in self.list_approachs}
        self.dic_entering_approach_lanes = {"W": [0], "E": [0], "S": [0], "N": [0]}
        self.dic_exiting_approach_lanes = {"W": [0], "E": [0], "S": [0], "N": [0]}

        # grid settings
        self.length_lane = 300
        self.length_terminal = 50
        self.length_grid = 5
        self.num_grid = int(self.length_lane // self.length_grid)

        self.list_phases = dic_traffic_env_conf["PHASE"][dic_traffic_env_conf['SIMULATOR_TYPE']]


        # generate all lanes
        self.list_entering_lanes = []
        for approach in self.list_approachs:
            self.list_entering_lanes += [self.dic_entering_approach_to_edge[approach] + '_' + str(i) for i in
                                         range(sum(list(dic_traffic_env_conf["LANE_NUM"].values())))]
        self.list_exiting_lanes = []
        for approach in self.list_approachs:
            self.list_exiting_lanes += [self.dic_exiting_approach_to_edge[approach] + '_' + str(i) for i in
                                        range(sum(list(dic_traffic_env_conf["LANE_NUM"].values())))]

        self.list_lanes = self.list_entering_lanes + self.list_exiting_lanes

        self.adjacency_row = light_id_dict['adjacency_row']
        # w_in
        self.in_adjacency_row = light_id_dict['in_adjacency_row']
        self.neighbor_ENWS = light_id_dict['neighbor_ENWS']
        self.neighbor_lanes_ENWS = light_id_dict['entering_lane_ENWS']

        def _get_top_k_lane(lane_id_list, top_k_input):
            top_k_lane_indexes = []
            for i in range(top_k_input):
                lane_id = lane_id_list[i] if i < len(lane_id_list) else None
                top_k_lane_indexes.append(lane_id)
            return top_k_lane_indexes

        self._adjacency_row_lanes = {}
        # _adjacency_row_lanes is the lane id, not index
        for lane_id in self.list_entering_lanes:
            if lane_id in light_id_dict['adjacency_matrix_lane']:
                self._adjacency_row_lanes[lane_id] = light_id_dict['adjacency_matrix_lane'][lane_id]
            else:
                self._adjacency_row_lanes[lane_id] = [_get_top_k_lane([], self.dic_traffic_env_conf["TOP_K_ADJACENCY_LANE"]),
                                                 _get_top_k_lane([], self.dic_traffic_env_conf["TOP_K_ADJACENCY_LANE"])]
        # order is the entering lane order, each element is list of two lists

        self.adjacency_row_lane_id_local = {}
        for index, lane_id in enumerate(self.list_entering_lanes):
            self.adjacency_row_lane_id_local[lane_id] = index

        # previous & current
        self.dic_lane_vehicle_previous_step = {}
        self.dic_lane_waiting_vehicle_count_previous_step = {}

        self.dic_vehicle_speed_previous_step = {}
        self.dic_vehicle_distance_previous_step = {}

        self.dic_lane_vehicle_current_step = {}
        self.dic_lane_waiting_vehicle_count_current_step = {}
        self.dic_vehicle_speed_current_step = {}
        self.dic_vehicle_distance_current_step = {}

        self.list_lane_vehicle_previous_step = []
        self.list_lane_vehicle_current_step = []

        # -1: all yellow, -2: all red, -3: none
        self.all_yellow_phase_index = -1
        self.all_red_phase_index = -2

        self.current_phase_index = 1
        self.previous_phase_index = 1
        self.eng.set_tl_phase(self.inter_name, self.current_phase_index)
        path_to_log_file = os.path.join(self.path_to_log, "signal_inter_{0}.txt".format(self.inter_name))
        df = [self.get_current_time(), self.current_phase_index]
        df = pd.DataFrame(df)
        df = df.transpose()
        df.to_csv(path_to_log_file, mode='a', header=False, index=False)

        self.next_phase_to_set_index = None
        self.current_phase_duration = -1
        self.all_red_flag = False
        self.all_yellow_flag = False
        self.flicker = 0

        self.dic_vehicle_min_speed = {}  # this second
        self.dic_vehicle_arrive_leave_time = dict()  # cumulative

        self.dic_feature = {}  # this second
        self.dic_feature_previous_step = {}  # this second

    def build_adjacency_row_lane(self, lane_id_to_global_index_dict):
        self.adjacency_row_lanes = [] # order is the entering lane order, each element is list of two lists
        for entering_lane_id in self.list_entering_lanes:
            _top_k_entering_lane, _top_k_leaving_lane = self._adjacency_row_lanes[entering_lane_id]
            top_k_entering_lane = []
            top_k_leaving_lane = []
            for lane_id in _top_k_entering_lane:
                top_k_entering_lane.append(lane_id_to_global_index_dict[lane_id] if lane_id is not None else -1)
            for lane_id in _top_k_leaving_lane:
                top_k_leaving_lane.append(lane_id_to_global_index_dict[lane_id]
                                          if (lane_id is not None) and (lane_id in lane_id_to_global_index_dict.keys())  # TODO leaving lanes of system will also have -1
                                          else -1)
            self.adjacency_row_lanes.append([top_k_entering_lane, top_k_leaving_lane])

    # set
    def set_signal(self, action, action_pattern, yellow_time, all_red_time):
        if self.all_yellow_flag:
            # in yellow phase
            self.flicker = 0
            if self.current_phase_duration >= yellow_time: # yellow time reached
                self.current_phase_index = self.next_phase_to_set_index
                self.eng.set_tl_phase(self.inter_name, self.current_phase_index) # if multi_phase, need more adjustment
                path_to_log_file = os.path.join(self.path_to_log, "signal_inter_{0}.txt".format(self.inter_name))
                df = [self.get_current_time(), self.current_phase_index]
                df = pd.DataFrame(df)
                df = df.transpose()
                df.to_csv(path_to_log_file, mode='a', header=False, index=False)
                self.all_yellow_flag = False
            else:
                pass
        else:
            # determine phase
            if action_pattern == "switch": # switch by order
                if action == 0: # keep the phase
                    self.next_phase_to_set_index = self.current_phase_index
                elif action == 1: # change to the next phase
                    self.next_phase_to_set_index = (self.current_phase_index + 1) % len(self.list_phases) # if multi_phase, need more adjustment
                else:
                    sys.exit("action not recognized\n action must be 0 or 1")

            elif action_pattern == "set": # set to certain phase
                self.next_phase_to_set_index = self.DIC_PHASE_MAP[action] # if multi_phase, need more adjustment

            # set phase
            if self.current_phase_index == self.next_phase_to_set_index: # the light phase keeps unchanged
                pass
            else: # the light phase needs to change
                # change to yellow first, and activate the counter and flag
                self.eng.set_tl_phase(self.inter_name, 0) # !!! yellow, tmp
                path_to_log_file = os.path.join(self.path_to_log, "signal_inter_{0}.txt".format(self.inter_name))
                df = [self.get_current_time(), self.current_phase_index]
                df = pd.DataFrame(df)
                df = df.transpose()
                df.to_csv(path_to_log_file, mode='a', header=False, index=False)
                #traci.trafficlights.setRedYellowGreenState(
                #    self.node_light, self.all_yellow_phase_str)
                self.current_phase_index = self.all_yellow_phase_index
                self.all_yellow_flag = True
                self.flicker = 1

    # update inner measurements
    def update_previous_measurements(self):
        self.previous_phase_index = self.current_phase_index

        self.dic_lane_vehicle_previous_step = self.dic_lane_vehicle_current_step
        self.dic_lane_waiting_vehicle_count_previous_step = self.dic_lane_waiting_vehicle_count_current_step
        self.dic_vehicle_speed_previous_step = self.dic_vehicle_speed_current_step
        self.dic_vehicle_distance_previous_step = self.dic_vehicle_distance_current_step


    def update_current_measurements_map(self, simulator_state):
        ## need change, debug in seeing format
        def _change_lane_vehicle_dic_to_list(dic_lane_vehicle):
            list_lane_vehicle = []

            for value in dic_lane_vehicle.values():
                list_lane_vehicle.extend(value)

            return list_lane_vehicle

        if self.current_phase_index == self.previous_phase_index:
            self.current_phase_duration += 1
        else:
            self.current_phase_duration = 1

        self.dic_lane_vehicle_current_step = {}
        self.dic_lane_waiting_vehicle_count_current_step = {}
        for lane in self.list_entering_lanes:
            self.dic_lane_vehicle_current_step[lane] = simulator_state["get_lane_vehicles"][lane]
            self.dic_lane_waiting_vehicle_count_current_step[lane] = simulator_state["get_lane_waiting_vehicle_count"][lane]

        for lane in self.list_exiting_lanes:
            self.dic_lane_vehicle_current_step[lane] = simulator_state["get_lane_vehicles"][lane]
            self.dic_lane_waiting_vehicle_count_current_step[lane] = simulator_state["get_lane_waiting_vehicle_count"][lane]

        self.dic_vehicle_speed_current_step = simulator_state['get_vehicle_speed']
        self.dic_vehicle_distance_current_step = simulator_state['get_vehicle_distance']

        # get vehicle list
        self.list_lane_vehicle_current_step = _change_lane_vehicle_dic_to_list(self.dic_lane_vehicle_current_step)
        self.list_lane_vehicle_previous_step = _change_lane_vehicle_dic_to_list(self.dic_lane_vehicle_previous_step)

        list_vehicle_new_arrive = list(set(self.list_lane_vehicle_current_step) - set(self.list_lane_vehicle_previous_step))
        list_vehicle_new_left = list(set(self.list_lane_vehicle_previous_step) - set(self.list_lane_vehicle_current_step))
        list_vehicle_new_left_entering_lane_by_lane = self._update_leave_entering_approach_vehicle()
        list_vehicle_new_left_entering_lane = []
        for l in list_vehicle_new_left_entering_lane_by_lane:
            list_vehicle_new_left_entering_lane += l

        # update vehicle arrive and left time
        self._update_arrive_time(list_vehicle_new_arrive)
        self._update_left_time(list_vehicle_new_left_entering_lane)

        # update vehicle minimum speed in history, # to be implemented
        #self._update_vehicle_min_speed()

        # update feature
        self._update_feature_map(simulator_state)


    def update_current_measurements(self):
        ## need change, debug in seeing format
        def _change_lane_vehicle_dic_to_list(dic_lane_vehicle):
            list_lane_vehicle = []

            for value in dic_lane_vehicle.values():
                list_lane_vehicle.extend(value)

            return list_lane_vehicle

        if self.current_phase_index == self.previous_phase_index:
            self.current_phase_duration += 1
        else:
            self.current_phase_duration = 1


        self.dic_lane_vehicle_current_step =[] # = self.eng.get_lane_vehicles()
        #not implement
        flow_tmp = self.eng.get_lane_vehicles()
        self.dic_lane_vehicle_current_step = {key: None for key in self.list_entering_lanes}
        for lane in self.list_entering_lanes:
            self.dic_lane_vehicle_current_step[lane] = flow_tmp[lane]

        self.dic_lane_waiting_vehicle_count_current_step = self.eng.get_lane_waiting_vehicle_count()
        self.dic_vehicle_speed_current_step = self.eng.get_vehicle_speed()
        self.dic_vehicle_distance_current_step = self.eng.get_vehicle_distance()

        # get vehicle list
        self.list_lane_vehicle_current_step = _change_lane_vehicle_dic_to_list(self.dic_lane_vehicle_current_step)
        self.list_lane_vehicle_previous_step = _change_lane_vehicle_dic_to_list(self.dic_lane_vehicle_previous_step)

        list_vehicle_new_arrive = list(set(self.list_lane_vehicle_current_step) - set(self.list_lane_vehicle_previous_step))
        list_vehicle_new_left = list(set(self.list_lane_vehicle_previous_step) - set(self.list_lane_vehicle_current_step))
        list_vehicle_new_left_entering_lane_by_lane = self._update_leave_entering_approach_vehicle()
        list_vehicle_new_left_entering_lane = []
        for l in list_vehicle_new_left_entering_lane_by_lane:
            list_vehicle_new_left_entering_lane += l

        # update vehicle arrive and left time
        self._update_arrive_time(list_vehicle_new_arrive)
        self._update_left_time(list_vehicle_new_left_entering_lane)

        # update vehicle minimum speed in history, # to be implemented
        #self._update_vehicle_min_speed()

        # update feature
        self._update_feature()



    def _update_leave_entering_approach_vehicle(self):

        list_entering_lane_vehicle_left = []

        # update vehicles leaving entering lane
        if not self.dic_lane_vehicle_previous_step:
            for lane in self.list_entering_lanes:
                list_entering_lane_vehicle_left.append([])
        else:
            last_step_vehicle_id_list = []
            current_step_vehilce_id_list = []
            for lane in self.list_entering_lanes:
                last_step_vehicle_id_list.extend(self.dic_lane_vehicle_previous_step[lane])
                current_step_vehilce_id_list.extend(self.dic_lane_vehicle_current_step[lane])

            list_entering_lane_vehicle_left.append(
                    list(set(last_step_vehicle_id_list) - set(current_step_vehilce_id_list))
                )

        return list_entering_lane_vehicle_left

    def _update_arrive_time(self, list_vehicle_arrive):

        ts = self.get_current_time()
        # get dic vehicle enter leave time
        for vehicle in list_vehicle_arrive:
            if vehicle not in self.dic_vehicle_arrive_leave_time:
                self.dic_vehicle_arrive_leave_time[vehicle] = \
                    {"enter_time": ts, "leave_time": np.nan}
            else:
                #print("vehicle: %s already exists in entering lane!"%vehicle)
                #sys.exit(-1)
                pass

    def _update_left_time(self, list_vehicle_left):

        ts = self.get_current_time()
        # update the time for vehicle to leave entering lane
        for vehicle in list_vehicle_left:
            try:
                self.dic_vehicle_arrive_leave_time[vehicle]["leave_time"] = ts
            except KeyError:
                print("vehicle not recorded when entering")
                sys.exit(-1)


    def update_neighbor_info(self, neighbors, dic_feature):
        # print(dic_feature)
        none_dic_feature = deepcopy(dic_feature)
        for key in none_dic_feature.keys():
            if none_dic_feature[key] is not None:
                if "cur_phase" in key:
                    none_dic_feature[key] = [1] * len(none_dic_feature[key])
                else:
                    none_dic_feature[key] = [0] * len(none_dic_feature[key])
            else:
                none_dic_feature[key] = None
        for i in range(len(neighbors)):
            neighbor = neighbors[i]
            example_dic_feature = {}
            if neighbor is None:
                example_dic_feature["cur_phase_{0}".format(i)] = none_dic_feature["cur_phase"]
                example_dic_feature["time_this_phase_{0}".format(i)] = none_dic_feature["time_this_phase"]
                example_dic_feature["lane_num_vehicle_{0}".format(i)] = none_dic_feature["lane_num_vehicle"]
            else:
                example_dic_feature["cur_phase_{0}".format(i)] = neighbor.dic_feature["cur_phase"]
                example_dic_feature["time_this_phase_{0}".format(i)] = neighbor.dic_feature["time_this_phase"]
                example_dic_feature["lane_num_vehicle_{0}".format(i)] = neighbor.dic_feature["lane_num_vehicle"]
            dic_feature.update(example_dic_feature)
        return dic_feature

    @staticmethod
    def _add_suffix_to_dict_key(target_dict, suffix):
        keys = list(target_dict.keys())
        for key in keys:
            target_dict[key+"_"+suffix] = target_dict.pop(key)
        return target_dict


    def _update_feature_map(self, simulator_state):

        dic_feature = dict()

        dic_feature["cur_phase"] = [self.current_phase_index]
        dic_feature["time_this_phase"] = [self.current_phase_duration]
        dic_feature["vehicle_position_img"] = None #self._get_lane_vehicle_position(self.list_entering_lanes)
        dic_feature["vehicle_speed_img"] = None #self._get_lane_vehicle_speed(self.list_entering_lanes)
        dic_feature["vehicle_acceleration_img"] = None
        dic_feature["vehicle_waiting_time_img"] = None #self._get_lane_vehicle_accumulated_waiting_time(self.list_entering_lanes)

        dic_feature["lane_num_vehicle"] = self._get_lane_num_vehicle(self.list_entering_lanes)
        dic_feature["pressure"] = None # [self._get_pressure()]

        if self.fast_compute:
            dic_feature["coming_vehicle"] = None
            dic_feature["leaving_vehicle"] = None
        else:
            dic_feature["coming_vehicle"] = self._get_coming_vehicles(simulator_state)
            dic_feature["leaving_vehicle"] = self._get_leaving_vehicles(simulator_state)



        dic_feature["lane_num_vehicle_been_stopped_thres01"] = None # self._get_lane_num_vehicle_been_stopped(0.1, self.list_entering_lanes)
        dic_feature["lane_num_vehicle_been_stopped_thres1"] = self._get_lane_num_vehicle_been_stopped(1, self.list_entering_lanes)
        dic_feature["lane_queue_length"] = None # self._get_lane_queue_length(self.list_entering_lanes)
        dic_feature["lane_num_vehicle_left"] = None
        dic_feature["lane_sum_duration_vehicle_left"] = None
        dic_feature["lane_sum_waiting_time"] = None #self._get_lane_sum_waiting_time(self.list_entering_lanes)
        dic_feature["terminal"] = None


        dic_feature["adjacency_matrix"] = self._get_adjacency_row() # TODO this feature should be a dict? or list of lists
        # w_in
        dic_feature["in_adjacency_matrix"] = self._get_in_adjacency_row()

        dic_feature["adjacency_matrix_lane"] = self._get_adjacency_row_lane() #row: entering_lane # columns: [inputlanes, outputlanes]

        dic_feature['connectivity'] = self._get_connectivity(self.neighbor_lanes_ENWS)

        self.dic_feature = dic_feature

    # ================= calculate features from current observations ======================

    def _get_adjacency_row(self):
        return self.adjacency_row

    # w_in
    def _get_in_adjacency_row(self):
        return self.in_adjacency_row

    def _get_adjacency_row_lane(self):
        return self.adjacency_row_lanes

    def lane_position_mapper(self, lane_pos, bins):
        lane_pos_np = np.array(lane_pos)
        digitized = np.digitize(lane_pos_np, bins)
        position_counter = [len(lane_pos_np[digitized == i]) for i in range(1, len(bins))]
        return position_counter

    def _get_coming_vehicles(self, simulator_state):
        ## TODO f vehicle position   eng.get_vehicle_distance()  ||  eng.get_lane_vehicles()

        coming_distribution = []
        ## dimension = num_lane*3*num_list_entering_lanes

        lane_vid_mapping_dict = simulator_state['get_lane_vehicles']
        vid_distance_mapping_dict = simulator_state['get_vehicle_distance']

        ## TODO LANE LENGTH = 300
        bins = np.linspace(0, 300, 4).tolist()

        for lane in self.list_entering_lanes:
            coming_vehicle_position = []
            vehicle_position_lane = lane_vid_mapping_dict[lane]
            for vehicle in vehicle_position_lane:
                coming_vehicle_position.append(vid_distance_mapping_dict[vehicle])
            coming_distribution.extend(self.lane_position_mapper(coming_vehicle_position, bins))


        return coming_distribution

    def _get_leaving_vehicles(self, simulator_state):
        leaving_distribution = []
        ## dimension = num_lane*3*num_list_entering_lanes

        lane_vid_mapping_dict = simulator_state['get_lane_vehicles']
        vid_distance_mapping_dict = simulator_state['get_vehicle_distance']

        ## TODO LANE LENGTH = 300
        bins = np.linspace(0, 300, 4).tolist()

        for lane in self.list_exiting_lanes:
            coming_vehicle_position = []
            vehicle_position_lane = lane_vid_mapping_dict[lane]
            for vehicle in vehicle_position_lane:
                coming_vehicle_position.append(vid_distance_mapping_dict[vehicle])
            leaving_distribution.extend(self.lane_position_mapper(coming_vehicle_position, bins))

        return leaving_distribution

    def _get_pressure(self):
        ##TODO eng.get_vehicle_distance(), another way to calculate pressure & queue length

        pressure = 0
        all_enter_car_queue = 0
        for lane in self.list_entering_lanes:
            all_enter_car_queue += self.dic_lane_waiting_vehicle_count_current_step[lane]

        all_leaving_car_queue = 0
        for lane in self.list_exiting_lanes:
            all_leaving_car_queue += self.dic_lane_waiting_vehicle_count_current_step[lane]

        p = all_enter_car_queue - all_leaving_car_queue

        if p < 0:
            p = -p

        return p

    def _get_lane_queue_length(self, list_lanes):
        '''
        queue length for each lane
        '''
        return [self.dic_lane_waiting_vehicle_count_current_step[lane] for lane in list_lanes]

    def _get_lane_num_vehicle(self, list_lanes):
        '''
        vehicle number for each lane
        '''
        return [len(self.dic_lane_vehicle_current_step[lane]) for lane in list_lanes]

    def _get_connectivity(self, dic_of_list_lanes):
        '''
        vehicle number for each lane
        '''
        result = []
        for i in range(len(dic_of_list_lanes['lane_ids'])):
            num_of_vehicles_on_road = sum([len(self.dic_lane_vehicle_current_step[lane]) for lane in dic_of_list_lanes['lane_ids'][i]])
            result.append(num_of_vehicles_on_road)

        lane_length = [0] + dic_of_list_lanes['lane_length']
        if np.sum(result)==0:
            result=[1]+result 
        else:
            result = [np.sum(result)]+ result
        connectivity = list(np.array(result * np.exp(-np.array(lane_length)/(self.length_lane*4))))
        # print(connectivity)
        # sys.exit()
        return connectivity

    def _get_lane_sum_waiting_time(self, list_lanes):
        '''
        waiting time for each lane
        '''
        raise NotImplementedError

    def _get_lane_list_vehicle_left(self, list_lanes):
        '''
        get list of vehicles left at each lane
        ####### need to check
        '''

        raise NotImplementedError

    # non temporary
    def _get_lane_num_vehicle_left(self, list_lanes):

        list_lane_vehicle_left = self._get_lane_list_vehicle_left(list_lanes)
        list_lane_num_vehicle_left = [len(lane_vehicle_left) for lane_vehicle_left in list_lane_vehicle_left]
        return list_lane_num_vehicle_left

    def _get_lane_sum_duration_vehicle_left(self, list_lanes):

        ## not implemented error
        raise NotImplementedError

    def _get_lane_num_vehicle_been_stopped(self, thres, list_lanes):
        return [self.dic_lane_waiting_vehicle_count_current_step[lane] for lane in list_lanes]

    def _get_position_grid_along_lane(self, vec):
        pos = int(self.dic_vehicle_sub_current_step[vec][get_traci_constant_mapping("VAR_LANEPOSITION")])
        return min(pos//self.length_grid, self.num_grid)

    def _get_lane_vehicle_position(self, list_lanes):

        list_lane_vector = []
        for lane in list_lanes:
            lane_vector = np.zeros(self.num_grid)
            list_vec_id = self.dic_lane_vehicle_current_step[lane]
            for vec in list_vec_id:
                pos = int(self.dic_vehicle_distance_current_step[vec])
                pos_grid = min(pos//self.length_grid, self.num_grid)
                lane_vector[pos_grid] = 1
            list_lane_vector.append(lane_vector)
        return np.array(list_lane_vector)
    
    # debug
    def _get_vehicle_info(self, veh_id):
        try:
            pos = self.dic_vehicle_distance_current_step[veh_id]
            speed = self.dic_vehicle_speed_current_step[veh_id]
            return pos, speed
        except:
            return None, None


    def _get_lane_vehicle_speed(self, list_lanes):
        return [self.dic_vehicle_speed_current_step[lane] for lane in list_lanes]

    def _get_lane_vehicle_accumulated_waiting_time(self, list_lanes):

        raise NotImplementedError

    # ================= get functions from outside ======================
    def get_current_time(self):
        return self.eng.get_current_time()

    def get_dic_vehicle_arrive_leave_time(self):

        return self.dic_vehicle_arrive_leave_time

    def get_feature(self):

        return self.dic_feature

    def get_state(self, list_state_features):
        # customize your own state
        # print(list_state_features)
        # print(self.dic_feature)
        dic_state = {state_feature_name: self.dic_feature[state_feature_name] for state_feature_name in list_state_features}

        return dic_state

    def get_reward(self, dic_reward_info):
        # customize your own reward
        dic_reward = dict()
        dic_reward["flickering"] = None
        dic_reward["sum_lane_queue_length"] = None
        dic_reward["sum_lane_wait_time"] = None
        dic_reward["sum_lane_num_vehicle_left"] = None
        dic_reward["sum_duration_vehicle_left"] = None
        dic_reward["sum_num_vehicle_been_stopped_thres01"] = None
        dic_reward["sum_num_vehicle_been_stopped_thres1"] = np.sum(self.dic_feature["lane_num_vehicle_been_stopped_thres1"])

        dic_reward['pressure'] = None # np.sum(self.dic_feature["pressure"])

        reward = 0
        for r in dic_reward_info:
            if dic_reward_info[r] != 0:
                reward += dic_reward_info[r] * dic_reward[r]
        return reward

class AnonEnv:
    list_intersection_id = [
        "intersection_1_1"
    ]

    def __init__(self, path_to_log, path_to_work_directory, dic_traffic_env_conf):
        self.path_to_log = path_to_log
        self.path_to_work_directory = path_to_work_directory
        self.dic_traffic_env_conf = dic_traffic_env_conf
        self.simulator_type = self.dic_traffic_env_conf["SIMULATOR_TYPE"]

        self.list_intersection = None
        self.list_inter_log = None
        self.list_lanes = None
        self.system_states = None
        self.feature_name_for_neighbor = self._reduce_duplicates(self.dic_traffic_env_conf["LIST_STATE_FEATURE"])

        # check min action time
        if self.dic_traffic_env_conf["MIN_ACTION_TIME"] <= self.dic_traffic_env_conf["YELLOW_TIME"]:
            print ("MIN_ACTION_TIME should include YELLOW_TIME")
            pass
            #raise ValueError

        # touch new inter_{}.pkl (if exists, remove)
        for inter_ind in range(self.dic_traffic_env_conf["NUM_INTERSECTIONS"]):
            path_to_log_file = os.path.join(self.path_to_log, "inter_{0}.pkl".format(inter_ind))
            f = open(path_to_log_file, "wb")
            f.close()

    def reset(self):

        print("# self.eng.reset() to be implemented")

        cityflow_config = {
            "interval": self.dic_traffic_env_conf["INTERVAL"],
            "seed": 0,
            "laneChange": False,
            "dir": self.path_to_work_directory+"/",
            "roadnetFile": self.dic_traffic_env_conf["ROADNET_FILE"],
            "flowFile": self.dic_traffic_env_conf["TRAFFIC_FILE"],
            "rlTrafficLight": self.dic_traffic_env_conf["RLTRAFFICLIGHT"],
            "saveReplay": self.dic_traffic_env_conf["SAVEREPLAY"],
            "roadnetLogFile": "frontend/web/roadnetLogFile.json",
            "replayLogFile": "frontend/web/replayLogFile.txt"
        }
        print("=========================")
        print(cityflow_config)

        with open(os.path.join(self.path_to_work_directory,"cityflow.config"), "w") as json_file:
            json.dump(cityflow_config, json_file)
        self.eng = engine.Engine(os.path.join(self.path_to_work_directory,"cityflow.config"), thread_num=1)
        # self.load_roadnet()
        # self.load_flow()


        # get adjacency
        if self.dic_traffic_env_conf["USE_LANE_ADJACENCY"]:
            self.traffic_light_node_dict = self._adjacency_extraction_lane()
        else:
            self.traffic_light_node_dict = self._adjacency_extraction()
        # w_in
        if not self.dic_traffic_env_conf['ADJACENCY_BY_CONNECTION_OR_GEO']:
            self._get_in_adjacency_extraction()
        else:
            for i in self.traffic_light_node_dict.keys():
                raw_total_adjs = np.copy(self.traffic_light_node_dict[i]['adjacency_row'])
                raw_total_adjs[raw_total_adjs==-1] = 1e8
                raw_total_adjs=np.sort(raw_total_adjs,axis=-1)
                raw_total_adjs[raw_total_adjs==1e8] = -1
                # still put all -1 at the tail
                self.traffic_light_node_dict[i]['in_adjacency_row'] = raw_total_adjs


        # initialize intersections (grid)
        self.list_intersection = [Intersection((i+1, j+1), self.dic_traffic_env_conf, self.eng,
                                               self.traffic_light_node_dict["intersection_{0}_{1}".format(i+1, j+1)],self.path_to_log)
                                  for i in range(self.dic_traffic_env_conf["NUM_ROW"])
                                  for j in range(self.dic_traffic_env_conf["NUM_COL"])]

        self.list_inter_log = [[] for i in range(self.dic_traffic_env_conf["NUM_ROW"] *
                                                 self.dic_traffic_env_conf["NUM_COL"])]

        # set index for intersections and global index for lanes
        self.id_to_index = {}
        count_inter = 0
        for i in range(self.dic_traffic_env_conf["NUM_ROW"]):
            for j in range(self.dic_traffic_env_conf["NUM_COL"]):
                self.id_to_index['intersection_{0}_{1}'.format(i+1, j+1)] = count_inter
                count_inter += 1

        self.lane_id_to_index = {}
        count_lane = 0
        for i in range(len(self.list_intersection)): # TODO
            for j in range(len(self.list_intersection[i].list_entering_lanes)):
                lane_id = self.list_intersection[i].list_entering_lanes[j]
                if lane_id not in self.lane_id_to_index.keys():
                    self.lane_id_to_index[lane_id] = count_lane
                    count_lane += 1

        # build adjacency_matrix_lane in index from _adjacency_matrix_lane
        for inter in self.list_intersection:
            inter.build_adjacency_row_lane(self.lane_id_to_index)


        # get new measurements
        system_state_start_time = time.time()
        if self.dic_traffic_env_conf["FAST_COMPUTE"]:
            self.system_states = {"get_lane_vehicles": self.eng.get_lane_vehicles(),
                              "get_lane_waiting_vehicle_count": self.eng.get_lane_waiting_vehicle_count(),
                              "get_vehicle_speed": None,
                              "get_vehicle_distance": None
                              }
        else:
            self.system_states = {"get_lane_vehicles": self.eng.get_lane_vehicles(),
                              "get_lane_waiting_vehicle_count": self.eng.get_lane_waiting_vehicle_count(),
                              "get_vehicle_speed": self.eng.get_vehicle_speed(),
                              "get_vehicle_distance": self.eng.get_vehicle_distance()
                              }
        print("Get system state time: ", time.time()-system_state_start_time)

        update_start_time = time.time()
        for inter in self.list_intersection:
            inter.update_current_measurements_map(self.system_states)
        print("Update_current_measurements_map time: ", time.time()-update_start_time)

        #update neighbor's info
        neighbor_start_time = time.time()
        if self.dic_traffic_env_conf["NEIGHBOR"]:
            for inter in self.list_intersection:
                neighbor_inter_ids = inter.neighbor_ENWS
                neighbor_inters = []
                for neighbor_inter_id in neighbor_inter_ids:
                    if neighbor_inter_id is not None:
                        neighbor_inters.append(self.list_intersection[self.id_to_index[neighbor_inter_id]])
                    else:
                        neighbor_inters.append(None)
                inter.dic_feature = inter.update_neighbor_info(neighbor_inters,deepcopy(inter.dic_feature))
        print("Update_neighbor time: ", time.time()-neighbor_start_time)

        state, done = self.get_state()
        # print(state)
        return state

    def step(self, action):
        step_start_time = time.time()
        list_action_in_sec = [action]
        list_action_in_sec_display = [action]
        for i in range(self.dic_traffic_env_conf["MIN_ACTION_TIME"]-1):
            if self.dic_traffic_env_conf["ACTION_PATTERN"] == "switch":
                list_action_in_sec.append(np.zeros_like(action).tolist())
            elif self.dic_traffic_env_conf["ACTION_PATTERN"] == "set":
                list_action_in_sec.append(np.copy(action).tolist())
            list_action_in_sec_display.append(np.full_like(action, fill_value=-1).tolist())

        average_reward_action_list = [0]*len(action)
        for i in range(self.dic_traffic_env_conf["MIN_ACTION_TIME"]):

            action_in_sec = list_action_in_sec[i]
            action_in_sec_display = list_action_in_sec_display[i]

            instant_time = self.get_current_time()
            self.current_time = self.get_current_time()

            before_action_feature = self.get_feature()
            # state = self.get_state()

            if self.dic_traffic_env_conf['DEBUG']:
                print("time: {0}".format(instant_time))
            else:
                if i == 0:
                    print("time: {0}".format(instant_time))

            self._inner_step(action_in_sec)


            # get reward
            if self.dic_traffic_env_conf['DEBUG']:
                start_time = time.time()

            reward = self.get_reward()

            if self.dic_traffic_env_conf['DEBUG']:
                print("Reward time: {}".format(time.time()-start_time))


            for j in range(len(reward)):
                average_reward_action_list[j] = (average_reward_action_list[j] * i + reward[j]) / (i + 1)

            # average_reward_action = (average_reward_action*i + reward[0])/(i+1)

            # log
            self.log(cur_time=instant_time, before_action_feature=before_action_feature, action=action_in_sec_display)

            next_state, done = self.get_state()

        print("Step time: ", time.time() - step_start_time)
        return next_state, reward, done, average_reward_action_list

    # w_in
    def step_with_w_in(self, action, w_in):
        step_start_time = time.time()
        list_action_in_sec = [action]
        list_action_in_sec_display = [action]
        # w_in
        list_w_in_in_sec_display = [w_in.tolist()]
        for i in range(self.dic_traffic_env_conf["MIN_ACTION_TIME"]-1):
            if self.dic_traffic_env_conf["ACTION_PATTERN"] == "switch":
                list_action_in_sec.append(np.zeros_like(action).tolist())
            elif self.dic_traffic_env_conf["ACTION_PATTERN"] == "set":
                list_action_in_sec.append(np.copy(action).tolist())
            list_action_in_sec_display.append(np.full_like(action, fill_value=-1).tolist())
            # w_in
            list_w_in_in_sec_display.append(np.full_like(w_in, fill_value=-1).tolist())

        average_reward_action_list = [0]*len(action)
        for i in range(self.dic_traffic_env_conf["MIN_ACTION_TIME"]):

            action_in_sec = list_action_in_sec[i]
            action_in_sec_display = list_action_in_sec_display[i]
            # w_in
            w_in_in_sec_display = list_w_in_in_sec_display[i]

            instant_time = self.get_current_time()
            self.current_time = self.get_current_time()

            before_action_feature = self.get_feature()
            # state = self.get_state()

            if self.dic_traffic_env_conf['DEBUG']:
                print("time: {0}".format(instant_time))
            else:
                if i == 0:
                    print("time: {0}".format(instant_time))

            self._inner_step(action_in_sec)


            # get reward
            if self.dic_traffic_env_conf['DEBUG']:
                start_time = time.time()

            reward = self.get_reward()

            if self.dic_traffic_env_conf['DEBUG']:
                print("Reward time: {}".format(time.time()-start_time))


            for j in range(len(reward)):
                average_reward_action_list[j] = (average_reward_action_list[j] * i + reward[j]) / (i + 1)

            # average_reward_action = (average_reward_action*i + reward[0])/(i+1)

            # log
            self.log_with_w_in(cur_time=instant_time, before_action_feature=before_action_feature, action=action_in_sec_display,
                w_in=w_in_in_sec_display)

            next_state, done = self.get_state()

        print("Step time: ", time.time() - step_start_time)
        return next_state, reward, done, average_reward_action_list

    def _inner_step(self, action):

        # copy current measurements to previous measurements
        for inter in self.list_intersection:
            inter.update_previous_measurements()

        # set signals
        # multi_intersection decided by action {inter_id: phase}
        for inter_ind, inter in enumerate(self.list_intersection):
            inter.set_signal(
                action=action[inter_ind],
                action_pattern=self.dic_traffic_env_conf["ACTION_PATTERN"],
                yellow_time=self.dic_traffic_env_conf["YELLOW_TIME"],
                all_red_time=self.dic_traffic_env_conf["ALL_RED_TIME"]
            )

        # run one step
        for i in range(int(1/self.dic_traffic_env_conf["INTERVAL"])):
            self.eng.next_step()

        if self.dic_traffic_env_conf['DEBUG']:
            start_time = time.time()

        system_state_start_time = time.time()
        if self.dic_traffic_env_conf["FAST_COMPUTE"]:
            self.system_states = {"get_lane_vehicles": self.eng.get_lane_vehicles(),
                              "get_lane_waiting_vehicle_count": self.eng.get_lane_waiting_vehicle_count(),
                              "get_vehicle_speed": None,
                              "get_vehicle_distance": None
                              }
        else:
            self.system_states = {"get_lane_vehicles": self.eng.get_lane_vehicles(),
                              "get_lane_waiting_vehicle_count": self.eng.get_lane_waiting_vehicle_count(),
                              "get_vehicle_speed": self.eng.get_vehicle_speed(),
                              "get_vehicle_distance": self.eng.get_vehicle_distance()
                              }

        # print("Get system state time: ", time.time()-system_state_start_time)

        if self.dic_traffic_env_conf['DEBUG']:
            print("Get system state time: {}".format(time.time()-start_time))
        # get new measurements

        if self.dic_traffic_env_conf['DEBUG']:
            start_time = time.time()

        update_start_time = time.time()
        for inter in self.list_intersection:
            inter.update_current_measurements_map(self.system_states)

        #update neighbor's info
        if self.dic_traffic_env_conf["NEIGHBOR"]:
            for inter in self.list_intersection:
                neighbor_inter_ids = inter.neighbor_ENWS
                neighbor_inters = []
                for neighbor_inter_id in neighbor_inter_ids:
                    if neighbor_inter_id is not None:
                        neighbor_inters.append(self.list_intersection[self.id_to_index[neighbor_inter_id]])
                    else:
                        neighbor_inters.append(None)
                inter.dic_feature = inter.update_neighbor_info(neighbor_inters, deepcopy(inter.dic_feature))


        # print("Update_current_measurements_map time: ", time.time()-update_start_time)

        if self.dic_traffic_env_conf['DEBUG']:
            print("Update measurements time: {}".format(time.time()-start_time))

        #self.log_lane_vehicle_position()
        # self.log_first_vehicle()
        #self.log_phase()

    def load_roadnet(self, roadnetFile=None):
        print("Start load roadnet")
        start_time = time.time()
        if not roadnetFile:
            roadnetFile = "roadnet_1_1.json"
        #print("/n/n", os.path.join(self.path_to_work_directory, roadnetFile))
        self.eng.load_roadnet(os.path.join(self.path_to_work_directory, roadnetFile))
        print("successfully load roadnet:{0}, time: {1}".format(roadnetFile,time.time()-start_time))

    def load_flow(self, flowFile=None):
        print("Start load flowFile")
        start_time = time.time()
        if not flowFile:
            flowFile = "flow_1_1.json"
        self.eng.load_flow(os.path.join(self.path_to_work_directory, flowFile))
        print("successfully load flowFile: {0}, time: {1}".format(flowFile, time.time()-start_time))

    def _check_episode_done(self, list_state):

        # ======== to implement ========

        return False

    @staticmethod
    def convert_dic_to_df(dic):
        list_df = []
        for key in dic:
            df = pd.Series(dic[key], name=key)
            list_df.append(df)
        return pd.DataFrame(list_df)

    def get_feature(self):
        list_feature = [inter.get_feature() for inter in self.list_intersection]
        return list_feature

    def get_state(self):
        # consider neighbor info
        list_state = [inter.get_state(self.dic_traffic_env_conf["LIST_STATE_FEATURE"]) for inter in self.list_intersection]
        done = self._check_episode_done(list_state)

        # print(list_state)

        return list_state, done

    @staticmethod
    def _reduce_duplicates(feature_name_list):
        new_list = set()
        for feature_name in feature_name_list:
            if feature_name[-1] in ["0","1","2","3"]:
                new_list.add(feature_name[:-2])
        return list(new_list)

    def get_reward(self):

        list_reward = [inter.get_reward(self.dic_traffic_env_conf["DIC_REWARD_INFO"]) for inter in self.list_intersection]

        return list_reward

    def get_current_time(self):
        return self.eng.get_current_time()

    def log(self, cur_time, before_action_feature, action):

        for inter_ind in range(len(self.list_intersection)):
            self.list_inter_log[inter_ind].append({"time": cur_time,
                                                    "state": before_action_feature[inter_ind],
                                                    "action": action[inter_ind]})

    # w_in
    def log_with_w_in(self, cur_time, before_action_feature, action, w_in):

        for inter_ind in range(len(self.list_intersection)):
            self.list_inter_log[inter_ind].append({"time": cur_time,
                                                    "state": before_action_feature[inter_ind],
                                                    "action": action[inter_ind],
                                                    "w_in": w_in[inter_ind]})


    def batch_log(self, start, stop):
        for inter_ind in range(start, stop):
            if int(inter_ind)%100 == 0:
                print("Batch log for inter ",inter_ind)
            path_to_log_file = os.path.join(self.path_to_log, "vehicle_inter_{0}.csv".format(inter_ind))
            dic_vehicle = self.list_intersection[inter_ind].get_dic_vehicle_arrive_leave_time()
            df = pd.DataFrame.from_dict(dic_vehicle,orient='index')
            df.to_csv(path_to_log_file, na_rep="nan")

            path_to_log_file = os.path.join(self.path_to_log, "inter_{0}.pkl".format(inter_ind))
            f = open(path_to_log_file, "wb")
            pickle.dump(self.list_inter_log[inter_ind], f)
            f.close()

    def bulk_log_multi_process(self, batch_size=100):
        assert len(self.list_intersection) == len(self.list_inter_log)
        if batch_size > len(self.list_intersection):
            batch_size_run = len(self.list_intersection)
        else:
            batch_size_run = batch_size
        process_list = []
        for batch in range(0, len(self.list_intersection), batch_size_run):
            start = batch
            stop = min(batch + batch_size, len(self.list_intersection))
            p = Process(target=self.batch_log, args=(start,stop))
            print("before")
            p.start()
            print("end")
            process_list.append(p)
        print("before join")

        for t in process_list:
            t.join()

        print("end join")

    def bulk_log(self):

        for inter_ind in range(len(self.list_intersection)):
            path_to_log_file = os.path.join(self.path_to_log, "vehicle_inter_{0}.csv".format(inter_ind))
            dic_vehicle = self.list_intersection[inter_ind].get_dic_vehicle_arrive_leave_time()
            df = self.convert_dic_to_df(dic_vehicle)
            df.to_csv(path_to_log_file, na_rep="nan")

        for inter_ind in range(len(self.list_inter_log)):
            path_to_log_file = os.path.join(self.path_to_log, "inter_{0}.pkl".format(inter_ind))
            f = open(path_to_log_file, "wb")
            pickle.dump(self.list_inter_log[inter_ind], f)
            f.close()

        self.eng.print_log(os.path.join(self.path_to_log, self.dic_traffic_env_conf["ROADNET_FILE"]),
                           os.path.join(self.path_to_log, "replay_1_1.txt"))

        #print("log files:", os.path.join("data", "frontend", "roadnet_1_1_test.json"))

    def log_attention(self, attention_dict):
        path_to_log_file = os.path.join(self.path_to_log, "attention.pkl")
        f = open(path_to_log_file, "wb")
        pickle.dump(attention_dict, f)
        f.close()

    def log_hidden_state(self, hidden_states):
        path_to_log_file = os.path.join(self.path_to_log, "hidden_states.pkl")

        with open(path_to_log_file, "wb") as f:
            pickle.dump(hidden_states, f)

    def log_lane_vehicle_position(self):
        def list_to_str(alist):
            new_str = ""
            for s in alist:
                new_str = new_str + str(s) + " "
            return new_str
        dic_lane_map = {
            "road_0_1_0_0": "w",
            "road_2_1_2_0": "e",
            "road_1_0_1_0": "s",
            "road_1_2_3_0": "n"
        }
        for inter in self.list_intersection:
            for lane in inter.list_entering_lanes:
                print(str(self.get_current_time()) + ", " + lane + ", " + list_to_str(inter._get_lane_vehicle_position([lane])[0]),
                      file=open(os.path.join(self.path_to_log, "lane_vehicle_position_%s.txt"%dic_lane_map[lane]), "a"))

    def log_lane_vehicle_position(self):
        def list_to_str(alist):
            new_str = ""
            for s in alist:
                new_str = new_str + str(s) + " "
            return new_str
        dic_lane_map = {
            "road_0_1_0_0": "w",
            "road_2_1_2_0": "e",
            "road_1_0_1_0": "s",
            "road_1_2_3_0": "n"
        }
        for inter in self.list_intersection:
            for lane in inter.list_entering_lanes:
                print(str(self.get_current_time()) + ", " + lane + ", " + list_to_str(inter._get_lane_vehicle_position([lane])[0]),
                      file=open(os.path.join(self.path_to_log, "lane_vehicle_position_%s.txt"%dic_lane_map[lane]), "a"))

    def log_first_vehicle(self):
        _veh_id = "flow_0_"
        _veh_id_2 = "flow_2_"
        _veh_id_3 = "flow_4_"
        _veh_id_4 = "flow_6_"

        for inter in self.list_intersection:
            for i in range(100):
                veh_id = _veh_id + str(i)
                veh_id_2 = _veh_id_2 + str(i)
                pos, speed = inter._get_vehicle_info(veh_id)
                pos_2, speed_2 = inter._get_vehicle_info(veh_id_2)
                # print(i, veh_id, pos, veh_id_2, speed, pos_2, speed_2)
                if not os.path.exists(os.path.join(self.path_to_log, "first_vehicle_info_a")):
                    os.makedirs(os.path.join(self.path_to_log, "first_vehicle_info_a"))

                if not os.path.exists(os.path.join(self.path_to_log, "first_vehicle_info_b")):
                    os.makedirs(os.path.join(self.path_to_log, "first_vehicle_info_b"))

                if pos and speed:
                    print("%f, %f, %f" % (self.get_current_time(), pos, speed),
                          file=open(os.path.join(self.path_to_log, "first_vehicle_info_a", "first_vehicle_info_a_%d.txt" % i), "a"))
                if pos_2 and speed_2:
                    print("%f, %f, %f" % (self.get_current_time(), pos_2, speed_2),
                          file=open(os.path.join(self.path_to_log, "first_vehicle_info_b", "first_vehicle_info_b_%d.txt" % i), "a"))

                veh_id_3 = _veh_id_3 + str(i)
                veh_id_4 = _veh_id_4 + str(i)
                pos_3, speed_3 = inter._get_vehicle_info(veh_id_3)
                pos_4, speed_4 = inter._get_vehicle_info(veh_id_4)
                # print(i, veh_id, pos, veh_id_2, speed, pos_2, speed_2)
                if not os.path.exists(os.path.join(self.path_to_log, "first_vehicle_info_c")):
                    os.makedirs(os.path.join(self.path_to_log, "first_vehicle_info_c"))

                if not os.path.exists(os.path.join(self.path_to_log, "first_vehicle_info_d")):
                    os.makedirs(os.path.join(self.path_to_log, "first_vehicle_info_d"))

                if pos_3 and speed_3:
                    print("%f, %f, %f" % (self.get_current_time(), pos_3, speed_3),
                          file=open(
                              os.path.join(self.path_to_log, "first_vehicle_info_c", "first_vehicle_info_a_%d.txt" % i),
                              "a"))
                if pos_4 and speed_4:
                    print("%f, %f, %f" % (self.get_current_time(), pos_4, speed_4),
                          file=open(
                              os.path.join(self.path_to_log, "first_vehicle_info_d", "first_vehicle_info_b_%d.txt" % i),
                              "a"))

    def log_phase(self):
        for inter in self.list_intersection:
            print("%f, %f" % (self.get_current_time(), inter.current_phase_index),
                  file=open(os.path.join(self.path_to_log, "log_phase.txt"), "a"))

    def _adjacency_extraction(self):
        traffic_light_node_dict = {}
        file = os.path.join(self.path_to_work_directory, self.dic_traffic_env_conf["ROADNET_FILE"])
        with open('{0}'.format(file)) as json_data:
            net = json.load(json_data)
            # print(net)
            for inter in net['intersections']:
                if not inter['virtual']:
                    traffic_light_node_dict[inter['id']] = {'location': {'x': float(inter['point']['x']),
                                                                       'y': float(inter['point']['y'])},
                                                            "total_inter_num": None, 'adjacency_row': None,
                                                            # w_in
                                                            "in_adjacency_row": None,
                                                            "inter_id_to_index": None,
                                                            "neighbor_ENWS": None,
                                                            "entering_lane_ENWS": None,}


            top_k = self.dic_traffic_env_conf["TOP_K_ADJACENCY"]
            total_inter_num = len(traffic_light_node_dict.keys())
            inter_id_to_index = {}

            edge_id_dict = {}
            for road in net['roads']:
                if road['id'] not in edge_id_dict.keys():
                    edge_id_dict[road['id']] = {}
                edge_id_dict[road['id']]['from'] = road['startIntersection']
                edge_id_dict[road['id']]['to'] = road['endIntersection']
                edge_id_dict[road['id']]['num_of_lane'] = len(road['lanes'])
                edge_id_dict[road['id']]['length'] = np.sqrt(np.square(pd.DataFrame(road['points'])).sum(axis=1)).sum()


            index = 0
            for i in traffic_light_node_dict.keys():
                inter_id_to_index[i] = index
                index += 1


            for i in traffic_light_node_dict.keys():
                traffic_light_node_dict[i]['inter_id_to_index'] = inter_id_to_index
                traffic_light_node_dict[i]['neighbor_ENWS'] = []
                traffic_light_node_dict[i]['entering_lane_ENWS'] = {"lane_ids": [], "lane_length": []}
                for j in range(4):
                    road_id = i.replace("intersection", "road")+"_"+str(j)
                    neighboring_node = edge_id_dict[road_id]['to']
                    # calculate the neighboring intersections
                    if neighboring_node not in traffic_light_node_dict.keys(): # virtual node
                        traffic_light_node_dict[i]['neighbor_ENWS'].append(None)
                    else:
                        traffic_light_node_dict[i]['neighbor_ENWS'].append(neighboring_node)
                    # calculate the entering lanes ENWS
                    for key, value in edge_id_dict.items():
                        if value['from'] == neighboring_node and value['to'] == i:
                            neighboring_road = key

                            neighboring_lanes = []
                            for k in range(value['num_of_lane']):
                                neighboring_lanes.append(neighboring_road+"_{0}".format(k))

                            traffic_light_node_dict[i]['entering_lane_ENWS']['lane_ids'].append(neighboring_lanes)
                            traffic_light_node_dict[i]['entering_lane_ENWS']['lane_length'].append(value['length'])


            for i in traffic_light_node_dict.keys():
                location_1 = traffic_light_node_dict[i]['location']

                # TODO return with Top K results
                if not self.dic_traffic_env_conf['ADJACENCY_BY_CONNECTION_OR_GEO']: # use geo-distance
                    row = np.array([0]*total_inter_num)
                    # row = np.zeros((self.dic_traffic_env_conf["NUM_ROW"],self.dic_traffic_env_conf["NUM_col"]))
                    for j in traffic_light_node_dict.keys():
                        location_2 = traffic_light_node_dict[j]['location']
                        dist = AnonEnv._cal_distance(location_1,location_2)
                        row[inter_id_to_index[j]] = dist
                    if len(row) == top_k:
                        adjacency_row_unsorted = np.argpartition(row, -1)[:top_k].tolist()
                    elif len(row) > top_k:
                        adjacency_row_unsorted = np.argpartition(row, top_k)[:top_k].tolist()
                    else:
                        adjacency_row_unsorted = [k for k in range(total_inter_num)]
                    adjacency_row_unsorted.remove(inter_id_to_index[i])
                    traffic_light_node_dict[i]['adjacency_row'] = [inter_id_to_index[i]] + adjacency_row_unsorted
                else: # use connection infomation
                    traffic_light_node_dict[i]['adjacency_row'] = [inter_id_to_index[i]]
                    for j in traffic_light_node_dict[i]['neighbor_ENWS']: ## TODO
                        if j is not None:
                            traffic_light_node_dict[i]['adjacency_row'].append(inter_id_to_index[j])
                        else:
                            traffic_light_node_dict[i]['adjacency_row'].append(-1)


                traffic_light_node_dict[i]['total_inter_num'] = total_inter_num

        return traffic_light_node_dict

    # w_in
    def _get_in_adjacency_extraction(self):
        """
        out-neighbors to in-neighbors
        """

        total_in_neighbors = []
        num_agents = len(self.traffic_light_node_dict)
        full_adj_matrix = -np.ones((num_agents,num_agents))
        for inter_ind, i in enumerate(self.traffic_light_node_dict.keys()):
            raw_total_adjs=np.array(self.traffic_light_node_dict[i]['adjacency_row'], dtype=np.int)
            raw_total_adjs[raw_total_adjs==-1] = 1e8
            raw_total_adjs=np.sort(raw_total_adjs,axis=-1)
            raw_total_adjs[raw_total_adjs==1e8] = -1
            # still put all -1 at the tail
            out_neighbors = raw_total_adjs[raw_total_adjs > 0]
            full_adj_matrix[inter_ind,out_neighbors] = 1
            # illegal places remain -1

        total_in_neighbors = []
        full_adj_matrix = np.transpose(full_adj_matrix, (1,0))
        for inter_ind in range(num_agents):
            # in-neighbors might be more than limited
            in_neighbors = np.where(full_adj_matrix[inter_ind,:] > 0)[0]
            total_in_neighbors.append(in_neighbors.tolist())
        # padding -1
        max_len = np.max([len(l) for l in total_in_neighbors])
        padding_in_neighbors = list(map(lambda l:l + 
            [-1 for _ in range(max_len-len(l))], total_in_neighbors))
        padding_in_neighbors = np.array(padding_in_neighbors, dtype=np.int)
        
        for inter_ind, i in enumerate(self.traffic_light_node_dict.keys()):
            self.traffic_light_node_dict[i]['in_adjacency_row'] = padding_in_neighbors[inter_ind]

    def _adjacency_extraction_lane(self):
        traffic_light_node_dict = {}
        file = os.path.join(self.path_to_work_directory, self.dic_traffic_env_conf["ROADNET_FILE"])

        roadnet = RoadNet('{0}'.format(file))
        with open('{0}'.format(file)) as json_data:
            net = json.load(json_data)
            # print(net)
            for inter in net['intersections']:
                if not inter['virtual']:
                    traffic_light_node_dict[inter['id']] = {'location': {'x': float(inter['point']['x']),
                                                                       'y': float(inter['point']['y'])},
                                                            "total_inter_num": None, 'adjacency_row': None,
                                                            # w_in
                                                            'in_adjacency_row': None,
                                                            "inter_id_to_index": None,
                                                            "neighbor_ENWS": None,
                                                            "entering_lane_ENWS": None,
                                                            "total_lane_num": None, 'adjacency_matrix_lane': None,
                                                            "lane_id_to_index": None,
                                                            "lane_ids_in_intersction": []
                                                            }

            top_k = self.dic_traffic_env_conf["TOP_K_ADJACENCY"]
            top_k_lane = self.dic_traffic_env_conf["TOP_K_ADJACENCY_LANE"]
            total_inter_num = len(traffic_light_node_dict.keys())

            edge_id_dict = {}
            for road in net['roads']:
                if road['id'] not in edge_id_dict.keys():
                    edge_id_dict[road['id']] = {}
                edge_id_dict[road['id']]['from'] = road['startIntersection']
                edge_id_dict[road['id']]['to'] = road['endIntersection']
                edge_id_dict[road['id']]['num_of_lane'] = len(road['lanes'])
                edge_id_dict[road['id']]['length'] = np.sqrt(np.square(pd.DataFrame(road['points'])).sum(axis=1)).sum()


            # set inter id to index dict
            inter_id_to_index = {}
            index = 0
            for i in traffic_light_node_dict.keys():
                inter_id_to_index[i] = index
                index += 1

            # set the neighbor_ENWS nodes and entring_lane_ENWS for intersections
            for i in traffic_light_node_dict.keys():
                traffic_light_node_dict[i]['inter_id_to_index'] = inter_id_to_index
                traffic_light_node_dict[i]['neighbor_ENWS'] = []
                traffic_light_node_dict[i]['entering_lane_ENWS'] = {"lane_ids": [], "lane_length": []}
                for j in range(4):
                    road_id = i.replace("intersection", "road")+"_"+str(j)
                    neighboring_node = edge_id_dict[road_id]['to']
                    # calculate the neighboring intersections
                    if neighboring_node not in traffic_light_node_dict.keys(): # virtual node
                        traffic_light_node_dict[i]['neighbor_ENWS'].append(None)
                    else:
                        traffic_light_node_dict[i]['neighbor_ENWS'].append(neighboring_node)
                    # calculate the entering lanes ENWS
                    for key, value in edge_id_dict.items():
                        if value['from'] == neighboring_node and value['to'] == i:
                            neighboring_road = key

                            neighboring_lanes = []
                            for k in range(value['num_of_lane']):
                                neighboring_lanes.append(neighboring_road+"_{0}".format(k))

                            traffic_light_node_dict[i]['entering_lane_ENWS']['lane_ids'].append(neighboring_lanes)
                            traffic_light_node_dict[i]['entering_lane_ENWS']['lane_length'].append(value['length'])


            lane_id_dict = roadnet.net_lane_dict
            total_lane_num = len(lane_id_dict.keys())
            # output an adjacentcy matrix for all the intersections
            # each row stands for a lane id,
            # each column is a list with two elements: first is the lane's entering_lane_LSR, second is the lane's leaving_lane_LSR
            def _get_top_k_lane(lane_id_list, top_k_input):
                top_k_lane_indexes = []
                for i in range(top_k_input):
                    lane_id = lane_id_list[i] if i < len(lane_id_list) else None
                    top_k_lane_indexes.append(lane_id)
                return top_k_lane_indexes

            adjacency_matrix_lane = {}
            for i in lane_id_dict.keys(): # Todo lane_ids should be in an order
                adjacency_matrix_lane[i] = [_get_top_k_lane(lane_id_dict[i]['input_lanes'], top_k_lane),
                                            _get_top_k_lane(lane_id_dict[i]['output_lanes'], top_k_lane)]



            for i in traffic_light_node_dict.keys():
                location_1 = traffic_light_node_dict[i]['location']

                # TODO return with Top K results
                if not self.dic_traffic_env_conf['ADJACENCY_BY_CONNECTION_OR_GEO']: # use geo-distance
                    row = np.array([0]*total_inter_num)
                    # row = np.zeros((self.dic_traffic_env_conf["NUM_ROW"],self.dic_traffic_env_conf["NUM_col"]))
                    for j in traffic_light_node_dict.keys():
                        location_2 = traffic_light_node_dict[j]['location']
                        dist = AnonEnv._cal_distance(location_1,location_2)
                        row[inter_id_to_index[j]] = dist
                    if len(row) == top_k:
                        adjacency_row_unsorted = np.argpartition(row, -1)[:top_k].tolist()
                    elif len(row) > top_k:
                        adjacency_row_unsorted = np.argpartition(row, top_k)[:top_k].tolist()
                    else:
                        adjacency_row_unsorted = [k for k in range(total_inter_num)]
                    adjacency_row_unsorted.remove(inter_id_to_index[i])
                    traffic_light_node_dict[i]['adjacency_row'] = [inter_id_to_index[i]] + adjacency_row_unsorted
                else: # use connection infomation
                    traffic_light_node_dict[i]['adjacency_row'] = [inter_id_to_index[i]]
                    for j in traffic_light_node_dict[i]['neighbor_ENWS']: ## TODO
                        if j is not None:
                            traffic_light_node_dict[i]['adjacency_row'].append(inter_id_to_index[j])
                        else:
                            traffic_light_node_dict[i]['adjacency_row'].append(-1)


                traffic_light_node_dict[i]['total_inter_num'] = total_inter_num
                traffic_light_node_dict[i]['total_lane_num'] = total_lane_num
                traffic_light_node_dict[i]['adjacency_matrix_lane'] = adjacency_matrix_lane



        return traffic_light_node_dict



    @staticmethod
    def _cal_distance(loc_dict1, loc_dict2):
        a = np.array((loc_dict1['x'], loc_dict1['y']))
        b = np.array((loc_dict2['x'], loc_dict2['y']))
        return np.sqrt(np.sum((a-b)**2))

    def end_sumo(self):
        print("anon process end")
        pass

if __name__ == '__main__':
    dic_agent_conf = {
    "PRIORITY": True,
    "nan_code":True,
    "att_regularization":False,
    "rularization_rate":0.03,
    "LEARNING_RATE": 0.001,
    "SAMPLE_SIZE": 1000,
    "BATCH_SIZE": 20,
    "EPOCHS": 100,
    "UPDATE_Q_BAR_FREQ": 5,
    "UPDATE_Q_BAR_EVERY_C_ROUND": False,
    "GAMMA": 0.8,
    "MAX_MEMORY_LEN": 10000,
    "PATIENCE": 10,
    "D_DENSE": 20,
    "N_LAYER": 2,
    #special care for pretrain
    "EPSILON": 0.8,
    "EPSILON_DECAY": 0.95,
    "MIN_EPSILON": 0.2,

    "LOSS_FUNCTION": "mean_squared_error",
    "SEPARATE_MEMORY": False,
    "NORMAL_FACTOR": 20,
    "TRAFFIC_FILE": "cross.2phases_rou01_equal_450.xml",
        }

    dic_exp_conf = {
        "RUN_COUNTS": 3600,
        "MODEL_NAME": "STGAT",


        "ROADNET_FILE": "roadnet_{0}.json".format("3_3"),

        "NUM_ROUNDS": 200,
        "NUM_GENERATORS": 4,

        "MODEL_POOL": False,
        "NUM_BEST_MODEL": 3,

        "PRETRAIN_NUM_ROUNDS": 0,
        "PRETRAIN_NUM_GENERATORS": 15,

        "AGGREGATE": False,
        "PRETRAIN": False,
        "DEBUG": False,
        "EARLY_STOP": True
    }

    dic_traffic_env_conf  = {
        "ADJACENCY_BY_CONNECTION_OR_GEO": True,
        "USE_LANE_ADJACENCY": True,
        "TRAFFIC_FILE": "/mnt/RLSignal_general/records/test/anon_3_3_test/anon_3_3_700_1.0.json",
        "THREADNUM": 8,
        "SAVEREPLAY": False,
        "RLTRAFFICLIGHT": True,
        "INTERVAL": 1,
        "NUM_INTERSECTIONS": 9,
        "ACTION_PATTERN": "set",
        "MEASURE_TIME": 10,
        "MIN_ACTION_TIME": 10,
        "YELLOW_TIME": 5,
        "DEBUG": False,
        "BINARY_PHASE_EXPANSION": True,
        "FAST_COMPUTE": True,
        'NUM_AGENTS': 1,

        "NEIGHBOR": False,
        "MODEL_NAME": "STGAT",
        "SIMULATOR_TYPE": "anon",
        "TOP_K_ADJACENCY":9,
        "TOP_K_ADJACENCY_LANE": 6,




            "SAVEREPLAY": False,
            "NUM_ROW": 4,
            "NUM_COL": 3,

            "VOLUME": 300,
            "ROADNET_FILE": "roadnet_{0}.json".format("3_4"),

            "LIST_STATE_FEATURE": [
                "cur_phase",
                # "time_this_phase",
                # "vehicle_position_img",
                # "vehicle_speed_img",
                # "vehicle_acceleration_img",
                # "vehicle_waiting_time_img",
                "lane_num_vehicle",
                # "lane_num_vehicle_been_stopped_thres01",
                # "lane_num_vehicle_been_stopped_thres1",
                # "lane_queue_length",
                # "lane_num_vehicle_left",
                # "lane_sum_duration_vehicle_left",
                # "lane_sum_waiting_time",
                # "terminal",
                # "coming_vehicle",
                # "leaving_vehicle",
                # "pressure"

                # "adjacency_matrix",
                # "lane_queue_length",
                "adjacency_matrix_lane",
            ],

                "DIC_FEATURE_DIM": dict(
                    D_LANE_QUEUE_LENGTH=(4,),
                    D_LANE_NUM_VEHICLE=(4,),

                    D_COMING_VEHICLE = (12,),
                    D_LEAVING_VEHICLE = (12,),

                    D_LANE_NUM_VEHICLE_BEEN_STOPPED_THRES1=(4,),
                    D_CUR_PHASE=(1,),
                    D_NEXT_PHASE=(1,),
                    D_TIME_THIS_PHASE=(1,),
                    D_TERMINAL=(1,),
                    D_LANE_SUM_WAITING_TIME=(4,),
                    D_VEHICLE_POSITION_IMG=(4, 60,),
                    D_VEHICLE_SPEED_IMG=(4, 60,),
                    D_VEHICLE_WAITING_TIME_IMG=(4, 60,),

                    D_PRESSURE=(1,),

                    D_ADJACENCY_MATRIX=(2,),
                    D_ADJACENCY_MATRIX_LANE=(6,)

                ),

            "DIC_REWARD_INFO": {
                "flickering": 0,
                "sum_lane_queue_length": 0,
                "sum_lane_wait_time": 0,
                "sum_lane_num_vehicle_left": 0,
                "sum_duration_vehicle_left": 0,
                "sum_num_vehicle_been_stopped_thres01": 0,
                "sum_num_vehicle_been_stopped_thres1": -0.25,
                "pressure": 0  # -0.25
            },

            "LANE_NUM": {
                "LEFT": 1,
                "RIGHT": 1,
                "STRAIGHT": 1
            },

            "PHASE": {
                "sumo": {
                    0: [0, 1, 0, 1, 0, 0, 0, 0],# 'WSES',
                    1: [0, 0, 0, 0, 0, 1, 0, 1],# 'NSSS',
                    2: [1, 0, 1, 0, 0, 0, 0, 0],# 'WLEL',
                    3: [0, 0, 0, 0, 1, 0, 1, 0]# 'NLSL',
                },
                "anon": {
                    # 0: [0, 0, 0, 0, 0, 0, 0, 0],
                    1: [0, 1, 1, 0, 1, 1, 0, 0, 1, 0, 0, 1],# 'WSES',
                    2: [0, 0, 1, 0, 0, 1, 0, 1, 1, 0, 1, 1],# 'NSSS',
                    3: [1, 0, 1, 1, 0, 1, 0, 0, 1, 0, 0, 1],# 'WLEL',
                    4: [0, 0, 1, 0, 0, 1, 1, 0, 1, 1, 0, 1]# 'NLSL',
                    # 'WSWL',
                    # 'ESEL',
                    # 'WSES',
                    # 'NSSS',
                    # 'NSNL',
                    # 'SSSL',
                },
            }
        }

    dic_path= {
            "PATH_TO_MODEL": "/Users/Wingslet/PycharmProjects/RLSignal/model/test/anon_3_3_test",
            "PATH_TO_WORK_DIRECTORY": "records/test/jinan_3_4",

            "PATH_TO_DATA": "data/test/",
            "PATH_TO_ERROR": "error/test/"
        }
    path_to_log = os.path.join(dic_path["PATH_TO_WORK_DIRECTORY"], "train_round",
                                    "round_" + str(0), "generator_" + str(0))

    env = AnonEnv(path_to_log, dic_path["PATH_TO_WORK_DIRECTORY"], dic_traffic_env_conf)
    env.reset()
    print("finish")