2023,江端さんの技術メモ

以下のtest.csvファイルを、gunplotで3D表示させたくて、色々やっています。

test.csv
#lat,lng,datetime,distance,speed
35.681100,139.758600,2023-05-22 02:10:30,313.307785,37.596934
35.683300,139.759900,2023-05-22 02:11:00,271.347933,32.561752
35.685300,139.760900,2023-05-22 02:11:30,240.030143,28.803617
35.685400,139.761000,2023-05-22 02:12:00,14.325264,1.719032
35.685400,139.761000,2023-05-22 02:12:30,0.000000,0.000000
35.685400,139.761100,2023-05-22 02:13:00,9.031610,1.083793
35.685200,139.763500,2023-05-22 02:13:30,217.896760,26.147611
35.684700,139.765700,2023-05-22 02:14:00,206.328362,24.759403
35.684200,139.768000,2023-05-22 02:14:30,215.040983,25.804918
35.685400,139.768400,2023-05-22 02:15:00,138.238013,16.588562

で、まあ、こんなgpファイルを使って試していましたが、上手く動きませんでした。

# Function to parse datetime string into a numerical value
strptime_datetime(x) = strptime('%Y-%m-%d %H:%M:%S', x)

# Set the output terminal to a 3D plot (you can change the output format if needed)
set terminal pngcairo enhanced size 800,600

# Set the data file separator to a comma
set datafile separator ','

# Set the axis labels
set xlabel 'Longitude (lng)'
set ylabel 'Latitude (lat)'
set zlabel 'Date and Time (datetime)'

# Set the view to a 3D perspective
set view 50,30,1,1


# Get the minimum and maximum datetime values from the data
stats 'test.csv' using (strptime_datetime(stringcolumn(3))) nooutput
min_datetime = STATS_min
max_datetime = STATS_max

# Set the range for the Z axis (datetime)
set zrange [min_datetime:max_datetime]

# Plot the data using the specified columns
#splot 'test.csv' using 2:1:(timecolumn(stringcolumn(3))) with points pointtype 7 pointsize 1 title 'Data Points'
#splot 'test.csv' using 2:1:(timecolumn(stringcolumn(3))) with points pointtype 7 pointsize 1 title 'Data Points'
#splot 'test.csv' using 2:1:(timecolumn(stringcolumn(3))):(0) with points pointtype 7 pointsize 1 title 'Data Points'

splot 'test.csv' using 2:1:(timecolumn(stringcolumn(3)))

時間のパース(strptime_datetime(x) = strptime('%Y-%m-%d %H:%M:%S', x))がやっぱり上手く動きませんでした。

でまあ、しょうがないので、"2023-05-22 02:10:30"を秒数に変換するプログラムを作成しました。

/* "2023-05-22 02:10:30"を秒数に変換するプログラムを作成しました
   c:\Users\ebata\gnuplot\convert_datetime.go
*/
package main

import (
	"encoding/csv"
	"fmt"
	"os"
	"time"
)

func main() {
	// 入力ファイル名と出力ファイル名を指定
	inputFileName := "test.csv"
	outputFileName := "formatted_test.csv"

	// 入力ファイルを開く
	inputFile, err := os.Open(inputFileName)
	if err != nil {
		fmt.Println("ファイルを開けませんでした:", err)
		return
	}
	defer inputFile.Close()

	// 出力ファイルを作成または上書き
	outputFile, err := os.Create(outputFileName)
	if err != nil {
		fmt.Println("ファイルを作成できませんでした:", err)
		return
	}
	defer outputFile.Close()

	// CSVリーダーとライターを作成
	reader := csv.NewReader(inputFile)
	writer := csv.NewWriter(outputFile)

	// ヘッダーを読み込み、書き込み
	header, err := reader.Read()
	if err != nil {
		fmt.Println("ヘッダーを読み込めませんでした:", err)
		return
	}
	writer.Write(header)

	// データを読み込んで秒数に変換して書き込み
	for {
		record, err := reader.Read()
		if err != nil {
			break
		}

		// datetime列をパースして秒数に変換
		datetime := record[2] // datetime列のインデックスを確認してください
		parsedTime, err := time.Parse("2006-01-02 15:04:05", datetime)
		if err != nil {
			fmt.Println("日時をパースできませんでした:", err)
			return
		}
		seconds := parsedTime.Unix()

		// 秒数に変換した値を新しい列として書き込み
		record = append(record, fmt.Sprintf("%d", seconds))
		writer.Write(record)
	}

	// ライターをフラッシュしてクローズ
	writer.Flush()

	if err := writer.Error(); err != nil {
		fmt.Println("書き込みエラー:", err)
		return
	}

	fmt.Println("変換が完了しました。出力ファイル:", outputFileName)
}

で、こんなのができました。
formatted_test.csv

#lat,lng,datetime,distance,speed
35.681100,139.758600,2023-05-22 02:10:30,313.307785,37.596934,1684721430
35.683300,139.759900,2023-05-22 02:11:00,271.347933,32.561752,1684721460
35.685300,139.760900,2023-05-22 02:11:30,240.030143,28.803617,1684721490
35.685400,139.761000,2023-05-22 02:12:00,14.325264,1.719032,1684721520
35.685400,139.761000,2023-05-22 02:12:30,0.000000,0.000000,1684721550
35.685400,139.761100,2023-05-22 02:13:00,9.031610,1.083793,1684721580
35.685200,139.763500,2023-05-22 02:13:30,217.896760,26.147611,1684721610
35.684700,139.765700,2023-05-22 02:14:00,206.328362,24.759403,1684721640

スクリプトではなく、コマンドで一つづつ入れていきました。

gnuplot> set datafile separator ','   ← これ凄く重要
gnuplot> splot "formatted_test.csv" using 2:1:6
の結果は以下の通りでした。

エリア限定
# x軸の範囲を指定
set xrange [139.6119000:139.6312000]

# y軸の範囲を指定
set yrange [35.3627000:35.3737000] ←これだと、地図のイメージと逆転するので
set yrange [35.3737000:35.3627000]  ← y軸の範囲を逆転させる

上記のxrangeとyrangeの範囲を地図で示すと、

の範囲となる。
領域の範囲内でのトラッキングデータの様子

gnuplotでx軸とy軸の範囲を指定解除するには、以下のコマンドを使用する。

gnuplot> set xrange [*:*]
gnuplot> set yrange [*:*]

gnuplot> splot "2023-topLatLngCounts4-1.csv" using 2:1:3
gnuplot> set yrange [35.3737000:35.3627000]    ← 軸の大小を引っくり返す
gnuplot> splot "2023-topLatLngCounts4-1.csv" using 2:1:3 ← 点を表示
gnuplot> replot "2023-topLatLngCounts4-1.csv" using 2:1:3 with impulses ← 縦軸表示

ちなみに、文字列が入っているcsvだと表示されないことが多いようだ。

2023,江端さんの技術メモ

「交通流動量 パーソントリップ発生・集中量データ」のデータベースの作り方

サンプルファイル

F:\しゅらばしゅう\S05-a-10_SYUTO_GML

s05_xml2csv.zip (ダウンロードして解凍して下さい。sample1.xmlも入っています)

package main

import (
	"encoding/csv"
	"encoding/xml"
	"fmt"
	"os"
)

// XMLデータの構造体定義
type Dataset struct {
	Items []OccurredConcentratedTrafficVolumeOfPersonTrip `xml:"Occurred_ConcentratedTrafficVolumeOfPersonTrip"`
}

type OccurredConcentratedTrafficVolumeOfPersonTrip struct {
	ID                                       string `xml:"id,attr"`
	UrbanArea                                int    `xml:"urbanArea"`
	SurveyYear                               int    `xml:"surveyYear"`
	ConcentratedOccurrence                   int    `xml:"concentratedOccurrence"`
	ZoneCode                                 int    `xml:"zoneCode"`
	Railroad_NumberOfTripsForGoingToWork     int    `xml:"railroad_NumberOfTripsForGoingToWork"`
	Railroad_NumberOfTripsForGoingToSchool   int    `xml:"railroad_NumberOfTripsForGoingToSchool"`
	Railroad_NumberOfFreeTrips               int    `xml:"railroad_NumberOfFreeTrips"`
	Railroad_NumberOfBusinessTrips           int    `xml:"railroad_NumberOfBusinessTrips"`
	Railroad_NumberOfTripsForGoingHome       int    `xml:"railroad_NumberOfTripsForGoingHome"`
	Railroad_TotalNumberOfTrips              int    `xml:"railroad_TotalNumberOfTrips"`
	Bus_NumberOfTripsForGoingToWork          int    `xml:"bus_NumberOfTripsForGoingToWork"`
	Bus_NumberOfTripsForGoingToSchool        int    `xml:"bus_NumberOfTripsForGoingToSchool"`
	Bus_NumberOfFreeTrips                    int    `xml:"bus_NumberOfFreeTrips"`
	Bus_NumberOfBusinessTrips                int    `xml:"bus_NumberOfBusinessTrips"`
	Bus_NumberOfTripsForGoingHome            int    `xml:"bus_NumberOfTripsForGoingHome"`
	Bus_TotalNumberOfTrips                   int    `xml:"bus_TotalNumberOfTrips"`
	Automobile_NumberOfTripsForGoingToWork   int    `xml:"automobile_NumberOfTripsForGoingToWork"`
	Automobile_NumberOfTripsForGoingToSchool int    `xml:"automobile_NumberOfTripsForGoingToSchool"`
	Automobile_NumberOfFreeTrips             int    `xml:"automobile_NumberOfFreeTrips"`
	Automobile_NumberOfBusinessTrips         int    `xml:"automobile_NumberOfBusinessTrips"`
	Automobile_NumberOfTripsForGoingHome     int    `xml:"automobile_NumberOfTripsForGoingHome"`
	Automobile_TotalNumberOfTrips            int    `xml:"automobile_TotalNumberOfTrips"`
	Motorcycle_NumberOfTripsForGoingToWork   int    `xml:"motorcycle_NumberOfTripsForGoingToWork"`
	Motorcycle_NumberOfTripsForGoingToSchool int    `xml:"motorcycle_NumberOfTripsForGoingToSchool"`
	Motorcycle_NumberOfFreeTrips             int    `xml:"motorcycle_NumberOfFreeTrips"`
	Motorcycle_NumberOfBusinessTrips         int    `xml:"motorcycle_NumberOfBusinessTrips"`
	Motorcycle_NumberOfTripsForGoingHome     int    `xml:"motorcycle_NumberOfTripsForGoingHome"`
	Motorcycle_TotalNumberOfTrips            int    `xml:"motorcycle_TotalNumberOfTrips"`
	Walk_NumberOfTripsForGoingToWork         int    `xml:"walk_NumberOfTripsForGoingToWork"`
	Walk_NumberOfTripsForGoingToSchool       int    `xml:"walk_NumberOfTripsForGoingToSchool"`
	Walk_NumberOfFreeTrips                   int    `xml:"walk_NumberOfFreeTrips"`
	Walk_NumberOfBusinessTrips               int    `xml:"walk_NumberOfBusinessTrips"`
	Walk_NumberOfTripsForGoingHome           int    `xml:"walk_NumberOfTripsForGoingHome"`
	Walk_TotalNumberOfTrips                  int    `xml:"walk_TotalNumberOfTrips"`
	TotalNumberOfTrips                       int    `xml:"totalNumberOfTrips"`
}

func main() {
	// XMLファイルを読み込む
	xmlData, err := os.Open("sample1.xml")
	if err != nil {
		fmt.Println("Error opening XML file:", err)
		return
	}
	defer xmlData.Close()

	// XMLデータをデコード
	var dataset Dataset
	decoder := xml.NewDecoder(xmlData)
	if err := decoder.Decode(&dataset); err != nil {
		fmt.Println("Error decoding XML:", err)
		return
	}

	// CSVファイルにデータを書き込み
	csvFile, err := os.Create("output.csv")
	if err != nil {
		fmt.Println("Error creating CSV file:", err)
		return
	}
	defer csvFile.Close()

	csvWriter := csv.NewWriter(csvFile)
	defer csvWriter.Flush()

	// CSVヘッダを書き込み
	headers := []string{
		"ID", "UrbanArea", "SurveyYear", "ConcentratedOccurrence", "ZoneCode",
		"Railroad_NumberOfTripsForGoingToWork", "Railroad_NumberOfTripsForGoingToSchool", "Railroad_NumberOfFreeTrips",
		"Railroad_NumberOfBusinessTrips", "Railroad_NumberOfTripsForGoingHome", "Railroad_TotalNumberOfTrips",
		"Bus_NumberOfTripsForGoingToWork", "Bus_NumberOfTripsForGoingToSchool", "Bus_NumberOfFreeTrips",
		"Bus_NumberOfBusinessTrips", "Bus_NumberOfTripsForGoingHome", "Bus_TotalNumberOfTrips",
		"Automobile_NumberOfTripsForGoingToWork", "Automobile_NumberOfTripsForGoingToSchool", "Automobile_NumberOfFreeTrips",
		"Automobile_NumberOfBusinessTrips", "Automobile_NumberOfTripsForGoingHome", "Automobile_TotalNumberOfTrips",
		"Motorcycle_NumberOfTripsForGoingToWork", "Motorcycle_NumberOfTripsForGoingToSchool", "Motorcycle_NumberOfFreeTrips",
		"Motorcycle_NumberOfBusinessTrips", "Motorcycle_NumberOfTripsForGoingHome", "Motorcycle_TotalNumberOfTrips",
		"Walk_NumberOfTripsForGoingToWork", "Walk_NumberOfTripsForGoingToSchool", "Walk_NumberOfFreeTrips",
		"Walk_NumberOfBusinessTrips", "Walk_NumberOfTripsForGoingHome", "Walk_TotalNumberOfTrips",
		"TotalNumberOfTrips",
	}

	if err := csvWriter.Write(headers); err != nil {
		fmt.Println("Error writing CSV headers:", err)
		return
	}

	// データをCSVに書き込み
	//for _, person := range root.Persons {
	for _, item := range dataset.Items {

		// fmt.Println("pass1")
		// fmt.Println(item)

		//record := []string{item.ID, fmt.Sprintf("%d", item.UrbanArea), fmt.Sprintf("%d", item.SurveyYear), fmt.Sprintf("%d", item.ConcentratedOccurrence), fmt.Sprintf("%d", item.ZoneCode)}

		record := []string{
			item.ID,
			fmt.Sprintf("%d", item.UrbanArea),
			fmt.Sprintf("%d", item.SurveyYear),
			fmt.Sprintf("%d", item.ConcentratedOccurrence),
			fmt.Sprintf("%d", item.ZoneCode),
			fmt.Sprintf("%d", item.Railroad_NumberOfTripsForGoingToWork),
			fmt.Sprintf("%d", item.Railroad_NumberOfTripsForGoingToSchool),
			fmt.Sprintf("%d", item.Railroad_NumberOfFreeTrips),
			fmt.Sprintf("%d", item.Railroad_NumberOfBusinessTrips),
			fmt.Sprintf("%d", item.Railroad_NumberOfTripsForGoingHome),
			fmt.Sprintf("%d", item.Railroad_TotalNumberOfTrips),
			fmt.Sprintf("%d", item.Bus_NumberOfTripsForGoingToWork),
			fmt.Sprintf("%d", item.Bus_NumberOfTripsForGoingToSchool),
			fmt.Sprintf("%d", item.Bus_NumberOfFreeTrips),
			fmt.Sprintf("%d", item.Bus_NumberOfBusinessTrips),
			fmt.Sprintf("%d", item.Bus_NumberOfTripsForGoingHome),
			fmt.Sprintf("%d", item.Bus_TotalNumberOfTrips),
			fmt.Sprintf("%d", item.Automobile_NumberOfTripsForGoingToWork),
			fmt.Sprintf("%d", item.Automobile_NumberOfTripsForGoingToSchool),
			fmt.Sprintf("%d", item.Automobile_NumberOfFreeTrips),
			fmt.Sprintf("%d", item.Automobile_NumberOfBusinessTrips),
			fmt.Sprintf("%d", item.Automobile_NumberOfTripsForGoingHome),
			fmt.Sprintf("%d", item.Automobile_TotalNumberOfTrips),
			fmt.Sprintf("%d", item.Motorcycle_NumberOfTripsForGoingToWork),
			fmt.Sprintf("%d", item.Motorcycle_NumberOfTripsForGoingToSchool),
			fmt.Sprintf("%d", item.Motorcycle_NumberOfFreeTrips),
			fmt.Sprintf("%d", item.Motorcycle_NumberOfBusinessTrips),
			fmt.Sprintf("%d", item.Motorcycle_NumberOfTripsForGoingHome),
			fmt.Sprintf("%d", item.Motorcycle_TotalNumberOfTrips),
			fmt.Sprintf("%d", item.Walk_NumberOfTripsForGoingToWork),
			fmt.Sprintf("%d", item.Walk_NumberOfTripsForGoingToSchool),
			fmt.Sprintf("%d", item.Walk_NumberOfFreeTrips),
			fmt.Sprintf("%d", item.Walk_NumberOfBusinessTrips),
			fmt.Sprintf("%d", item.Walk_NumberOfTripsForGoingHome),
			fmt.Sprintf("%d", item.Walk_TotalNumberOfTrips),
			fmt.Sprintf("%d", item.TotalNumberOfTrips),
		}

		if err := csvWriter.Write(record); err != nil {
			fmt.Println("Error writing CSV record:", err)
			return
		}

	}

	fmt.Println("CSV file successfully created.")
}

 

2023,江端さんの技術メモ

マルチエージェントシミュレーションにおいて、エージェント間の相互作用を実現するためには、さまざまな方法が存在します。「模倣戦略」や「ペアワイズフェルミ法」以外の方法として以下のいくつかを考えることができます:

  1. ゲーム理論:
    • ナッシュ均衡やコアの概念を使用して、エージェント間の戦略的相互作用をモデル化する方法です。ゲーム理論を用いることで、競争や協力などの行動を表現できます。
  2. マーケットモデル:
    • エージェントがリソースや商品の交換を行う市場モデルを使用して、エージェント間の相互作用をシミュレートできます。価格設定や需要と供給の変動を考慮することができます。
  3. エージェントベースのモデリング:
    • エージェントが独自の行動ルールと目標を持つモデルを作成し、それらのエージェントが環境内で相互作用する方法です。エージェントの行動は、周囲のエージェントや環境に影響を与えます。
  4. 群集ダイナミクス:
    • エージェントが群れを形成し、群れ内での相互作用をモデル化する方法です。鳥の群れや魚の群れのような集団行動を研究するのに適しています。
  5. 進化アルゴリズム:
    • 遺伝的アルゴリズムや遺伝的プログラミングを使用して、エージェントの戦略や行動を進化させる方法です。最適な戦略を見つけるために遺伝子の操作を行います。

これらの方法は、異なる問題領域やシミュレーションの目的に応じて選択できます。具体的なシナリオや要件に合わせて、最適な方法を選んでください。

2023,江端さんの技術メモ

データベース(DB)のデータのインポートを開始して、すでに3時間が経過しようとしています。このインポートが、これまで作ってきたDBのテーブル全部を吹き飛ばすことがなかろうかと、とても心配しています。

User

dockerを使ってPostgresqlでDBを作っています。特にパラメタのセットとかはしていません。このような場合、DB容量の上限値をざっくりとで結構ですので、教えて下さい。

ChatGPT
Dockerコンテナ内でPostgreSQLをデフォルトの設定で実行する場合、データベースのサイズに特別な制限は設定されません。制限はDockerコンテナ自体のディスク容量に依存し、PostgreSQL自体は通常制限を設けません。

Dockerコンテナのディスク容量は、ホストマシンのディスク容量に依存し、Dockerコンテナが実行されているディスク領域によって決まります。したがって、デフォルトの設定であれば、PostgreSQLデータベースの容量制限はホストマシンのディスク容量に影響されます。

したがって、特定の容量制限を設定していない場合、PostgreSQLデータベースの上限容量は、ホストマシンのディスク容量に依存し、通常は制約を受けません。ホストマシンのディスク容量に制約がある場合、それが制限となります。

PostgreSQL自体にはデータベースサイズの制約を設定する設定がありますが、通常はデフォルトの設定で制限を設けることはありません。したがって、Dockerコンテナ内で特別な設定を行っていない場合、PostgreSQLデータベースのサイズは通常はホストマシンのディスク容量によって制約されます。


このような回答を得て、先ずは安堵していますが、それでもやっぱり怖いです。
Dockerコンテナの丸ごとバックアップを試みるべきでしょうか。
以前やったときは、データ本体がスッポリ抜けていたような、嫌な記憶があります。

 

 

2023,江端さんの技術メモ

Q: 緯度、軽度の点からなる鉄道路線のCSVデータがあるのですが、これが線の順番に並んでおらず、バラバラの順番になっています。このcsvファイルを線の順番に並び替えるGo言語プログラムを作成して下さい。

点画にすれば、こんな感じ

でも、線画にすると、こんな感じになる

で、このcsvのファイル名を、"your_csv_file.csv"として、以下のプログラムに放り込む

package main

import (
	"bufio"
	"encoding/csv"
	"fmt"
	"io"
	"math"
	"os"
	"strconv"
)

// 座標を表す構造体
type Coordinate struct {
	Latitude  float64
	Longitude float64
}

func main() {
	// CSVファイルのパス
	filePath := "your_csv_file.csv"

	// CSVファイルを読み込む
	coordinates, err := readCoordinatesFromFile(filePath)
	if err != nil {
		fmt.Println("Error reading CSV file:", err)
		return
	}

	// 座標を線の順番に並び替え
	sortedCoordinates := sortCoordinates(coordinates)

	// 結果を表示
	for _, coord := range sortedCoordinates {
		fmt.Printf("%f, %f\n", coord.Latitude, coord.Longitude)
	}
}

// CSVファイルから座標を読み込む関数
func readCoordinatesFromFile(filePath string) ([]Coordinate, error) {
	var coordinates []Coordinate

	file, err := os.Open(filePath)
	if err != nil {
		return nil, err
	}
	defer file.Close()

	reader := csv.NewReader(bufio.NewReader(file))
	for {
		line, err := reader.Read()
		if err == io.EOF {
			break
		} else if err != nil {
			return nil, err
		}

		// CSVの各行から緯度と経度を抽出
		latitude, err := strconv.ParseFloat(line[0], 64)
		if err != nil {
			return nil, err
		}
		longitude, err := strconv.ParseFloat(line[1], 64)
		if err != nil {
			return nil, err
		}

		coordinates = append(coordinates, Coordinate{Latitude: latitude, Longitude: longitude})
	}

	return coordinates, nil
}

// 座標を線の順番に並び替える関数
func sortCoordinates(coordinates []Coordinate) []Coordinate {
	// 最初の座標をスタート地点として選択
	startIndex := 0
	sortedCoordinates := []Coordinate{coordinates[startIndex]}
	coordinates = append(coordinates[:startIndex], coordinates[startIndex+1:]...)

	// 座標を距離に基づいてソート
	for len(coordinates) > 0 {
		minIndex := findNearestCoordinateIndex(coordinates, sortedCoordinates[len(sortedCoordinates)-1])
		sortedCoordinates = append(sortedCoordinates, coordinates[minIndex])
		coordinates = append(coordinates[:minIndex], coordinates[minIndex+1:]...)
	}

	return sortedCoordinates
}

// 最も距離の近い座標のインデックスを検索する関数
func findNearestCoordinateIndex(coordinates []Coordinate, reference Coordinate) int {
	minDistance := math.MaxFloat64
	minIndex := 0

	for i, coord := range coordinates {
		distance := calculateDistance(coord, reference)
		if distance < minDistance {
			minDistance = distance
			minIndex = i
		}
	}

	return minIndex
}

// Haversine式を使用して座標間の距離を計算する関数
func calculateDistance(coord1, coord2 Coordinate) float64 {
	earthRadius := 6371.0 // 地球の半径(キロメートル)

	// 度数法からラジアンに変換
	lat1 := degToRad(coord1.Latitude)
	lon1 := degToRad(coord1.Longitude)
	lat2 := degToRad(coord2.Latitude)
	lon2 := degToRad(coord2.Longitude)

	// Haversine式による距離計算
	dlon := lon2 - lon1
	dlat := lat2 - lat1
	a := math.Pow(math.Sin(dlat/2), 2) + math.Cos(lat1)*math.Cos(lat2)*math.Pow(math.Sin(dlon/2), 2)
	c := 2 * math.Atan2(math.Sqrt(a), math.Sqrt(1-a))
	distance := earthRadius * c

	return distance
}

// 度数法をラジアンに変換する関数
func degToRad(deg float64) float64 {
	return deg * (math.Pi / 180)
}

でてきたcsvをファイルにして表示すると、こんな感じになった。

だいぶ良くなったんだけど、変な直線が一本引かれている。

考察した結果、このプログラムは、
(Step.1) 出発点に一番近いノードを探す。
(Step.2)そのノードを次の出発点にして(Step.1)を行う
という処理をしているため、使われていないノードが、候補として残ってしまうという問題が生じる。

で、終端に至ったノードは使われていないノードの中から、一番近いノードを探してくるため、このような問題が発生してしまう、ということだろう。

対策は結構簡単で、ノード距離が異様に大きい場合は、そこで打ち切る、という処理を入れれば足る
(面倒なので、今回は、手動で削除した)

以上

2023,江端さんの技術メモ

で「エクスポートをクリック」 → map.osmがダウンロードされる

JOSMで表示する

フィルタを選択

Ctrl-aで選択

「編集」→「座標をコピー」で、座標コピー

ここで、エディター等に(test.csvなどとして)コピーする。

【QGIS】緯度・経度をもつCSVデータを読み込み、地図に表示する

を参照して、test.csvをQGISで表示する。

このように座標が取得できているのが確認できました。

根岸線を削除する場合、JOSMで場所を指定して丹念に削除していく

で、この座標さえ取れれば、

Go言語で、BusRouteの点で繋がれた線とみなして、trackingDataの各点から、その線からの最短距離を算出するようなプログラムの改造をお願いできますか?

が使えるようになる。

ダメ。これで取得した座標は、出現順番がバラバラになっているので、補正処理が必要となる。
今回の場合、南北方向にのみ伸びているので、緯度で並べかえれば良いが、そうでない場合は、別途処理が必要となると思うが(今回は必要ないので、これは省略)

この後であれば、上記のプログラムが使えるはず。

 

ちなみに、
osmosis --read-xml map.osm --tf accept-nodes railway=* --write-xml railway_nodes.osm
osmosis --read-xml map.osm --tf accept-relations railway=* --used-way --write-xml railway_data.osm

は、上手く動かせませんでした。

 

 

2023,江端さんの技術メモ

特別なネットワークツールを使わずに、ネットワークのトラフィックを計測する方法。
(WinDumpもWireSharkも面倒だなーと)

PowerShellから、これをコピペ(xxxx.batにする必要なし)すると、network_data.logというファイルができる

while ($true) {
    $NetworkInterfaces = Get-NetAdapter | Where-Object { $_.Status -eq 'Up' }
 
    foreach ($NetworkInterface in $NetworkInterfaces) {
        $Statistics = Get-NetAdapterStatistics -Name $NetworkInterface.Name
        $ReceivedBytes = $Statistics.ReceivedBytes
        $SentBytes = $Statistics.SentBytes
 
        $TimeStamp = Get-Date -Format "yyyy-MM-dd HH:mm:ss"
        $Record = "$TimeStamp, Interface: $($NetworkInterface.Name), Received Bytes: $ReceivedBytes bytes, Sent Bytes: $SentBytes bytes"
 
        Add-Content -Path "network_data.log" -Value $Record
    }
 
    Start-Sleep -Seconds 10  # 10秒待機
}

で、network_data.logの内容は、以下の通り

PS C:\Users\azureuser> more .\network_data.log
2023-12-07 10:02:07, Interface: Ethernet 7, Received Bytes: 182088626 bytes, Sent Bytes: 1971150950 bytes
2023-12-07 10:02:07, Interface: Ethernet, Received Bytes: 183155088 bytes, Sent Bytes: 1882163584 bytes
2023-12-07 10:02:17, Interface: Ethernet 7, Received Bytes: 182107502 bytes, Sent Bytes: 1971187408 bytes
2023-12-07 10:02:17, Interface: Ethernet, Received Bytes: 183176035 bytes, Sent Bytes: 1882198919 bytes
2023-12-07 10:02:27, Interface: Ethernet 7, Received Bytes: 182141183 bytes, Sent Bytes: 1971266248 bytes
2023-12-07 10:02:27, Interface: Ethernet, Received Bytes: 183207355 bytes, Sent Bytes: 1882275443 bytes

で、ここで、"Ethernet 7"と、"Ethernet"の2つのインターフェースがあることが分かるので、

$NetworkInterfaces = Get-NetAdapter | Where-Object { $_.Status -eq 'Up' -and $_.Name -eq 'Ethernet' }

とすると、'Ethernet' とか 'Ethernet 7'のどちらかを選ぶことができる。

2023,江端さんの技術メモ

ラズパイ4 + USBカメラ Webカメラ C270n HD 720P で、SRTストリームに変換してライブ映像を転送する方法(色々やってきたんだけど、この組み合わせだけがスポッと抜けていて、探すのに苦労した)

送信側
cam@cam-desktop:~$ gst-launch-1.0 v4l2src ! video/x-raw,format=YUY2,width=640,height=480,framerate=30/1 ! videoconvert ! x264enc tune=zerolatency ! mpegtsmux ! srtserversink uri=srt://:8888

受信側
cam@cam-desktop:~$ ffplay -i srt://localhost:8888

2023,江端さんの技術メモ

f:/しゅらばしゅう/有吉先生データ/Moove/2023-07/fit_track3.go
BusRouteの点で繋がれた線とみなして、trackingDataの各点から、その線からの最短距離を算出するようなプログラムの改造をお願いできますか?
はい、バス路線の各点を線とみなして、トラッキングデータの各点からその線までの最短距離を計算するプログラムの改造を行います。planar.DistanceFrom 関数を使用して、各トラッキングポイントからバス路線を形成する線への最短距離を計算できます。

 

package main

import (
	"fmt"

	"github.com/paulmach/orb"
	"github.com/paulmach/orb/planar"
)

// Point はデータ番号と緯度、経度を保持する構造体です
type Point struct {
	Number   int
	Lat, Lon float64
}

// トラッキングデータとバス路線データの変換
func convertToOrbPoint(p Point) orb.Point {
	return orb.Point{p.Lon, p.Lat}
}

func main() {
	// トラッキングデータ
	trackingData := []Point{
		{1, 139.6268272, 35.3715607},
		{2, 139.627135, 35.3719136},
		{3, 139.6274291, 35.3721591},
		{11, 139.6276967, 35.3724112},
		{101, 139.6276253, 35.372575},
		{59, 139.6274377, 35.3727348},
		{7, 139.6272077, 35.3728485},
		{99, 139.6263279, 35.3728985},
		{111, 139.6262229, 35.3728182},
		{131, 139.6260982, 35.3726421},
		{200, 139.6260349, 35.3724965},
		{32, 139.6257477, 35.3720127},
		{42, 139.6254685, 35.3705626},
		{44, 139.6254273, 35.3703156},
		{46, 139.62361383758, 35.36644117332},
	}

	// バス路線データ
	busRoute := []Point{
		{0, 139.6272077, 35.3728485},
		{0, 139.6263279, 35.3728985},
		{0, 139.6262229, 35.3728182},
		{0, 139.6260982, 35.3726421},
		{0, 139.6260349, 35.3724965},
		{0, 139.6257477, 35.3720127},
		{0, 139.6254685, 35.3705626},
		{0, 139.6254273, 35.3703156},
		{0, 139.62361383758, 35.36644117332},
		{0, 139.62386146588, 35.3665604971},
		{0, 139.62412597981, 35.36668746811},
		{0, 139.6245189, 35.3668106},
		{0, 139.6250347, 35.367043},
		{0, 139.62548981146, 35.36726113676},
	}

	// orbライブラリを使用して、トラッキングデータとバス路線データをorb.Pointのスライスに変換
	var trackingPoints []orb.Point
	var busRoutePoints []orb.Point
	for _, p := range trackingData {
		trackingPoints = append(trackingPoints, convertToOrbPoint(p))
	}
	for _, p := range busRoute {
		busRoutePoints = append(busRoutePoints, convertToOrbPoint(p))
	}

	// トラッキングポイントからバス路線上の最も近い点までの距離を計算し、データ番号とともに出力
	for _, trackPoint := range trackingData {
		distance := planar.DistanceFrom(orb.LineString(busRoutePoints), convertToOrbPoint(trackPoint))
		fmt.Printf("Tracking Point %d: Lat: %v, Lng: %v, Distance: %v meters\n", trackPoint.Number, trackPoint.Lat, trackPoint.Lon, distance)
	}
}

出力結果

Tracking Point 1: Lat: 139.6268272, Lng: 35.3715607, Distance: 0.001145488357906135 meters
Tracking Point 2: Lat: 139.627135, Lng: 35.3719136, Distance: 0.0009375188544507943 meters
Tracking Point 3: Lat: 139.6274291, Lng: 35.3721591, Distance: 0.0007240789459736531 meters
Tracking Point 11: Lat: 139.6276967, Lng: 35.3724112, Distance: 0.0006560124160327593 meters
Tracking Point 101: Lat: 139.6276253, Lng: 35.372575, Distance: 0.0004991913560911751 meters
Tracking Point 59: Lat: 139.6274377, Lng: 35.3727348, Distance: 0.000256569074508911 meters
Tracking Point 7: Lat: 139.6272077, Lng: 35.3728485, Distance: 0 meters
Tracking Point 99: Lat: 139.6263279, Lng: 35.3728985, Distance: 0 meters
Tracking Point 111: Lat: 139.6262229, Lng: 35.3728182, Distance: 0 meters
Tracking Point 131: Lat: 139.6260982, Lng: 35.3726421, Distance: 0 meters
Tracking Point 200: Lat: 139.6260349, Lng: 35.3724965, Distance: 0 meters
Tracking Point 32: Lat: 139.6257477, Lng: 35.3720127, Distance: 0 meters
Tracking Point 42: Lat: 139.6254685, Lng: 35.3705626, Distance: 0 meters
Tracking Point 44: Lat: 139.6254273, Lng: 35.3703156, Distance: 0 meters
Tracking Point 46: Lat: 139.62361383758, Lng: 35.36644117332, Distance: 0 meters

2023,江端さんの技術メモ

SELECT * FROM records WHERE start_at = '2023-03-20 05:03:00'; start_atの時間順に表示するSQLを教えて下さい。
SELECT * FROM records ORDER BY start_at ASC;

降順(新しいものから古いものへ)に並べ替えたい場合は、ASCDESC に変更します:
SELECT * FROM records ORDER BY start_at DESC;

select user_id, start_at, end_at from records order by start_at ASC where user_id = 7; のSQLは何が間違っていますか?
SELECT user_id, start_at, end_at FROM records WHERE user_id = 7 ORDER BY start_at ASC;

moove=# select place from records where id = 2; place ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- [{"lat": 34.7325, "lng": 135.4986}, {"lat": 34.7326, "lng": 135.4989}, {"lat": 34.7326, "lng": 135.4989}, {"lat": 34.7327, "lng": 135.4995}, {"lat": 34.7327, "lng": 135.5}, {"lat": 34.7328, "lng": 135.5002}, {"lat": 34.733, "lng": 135.5006}, {"lat": 34.7331, "lng": 135.501}, {"lat": 34.7334, "lng": 135.5015}, {"lat": 34.7336, "lng": 135.5018}, {"lat": 34.7336, "lng": 135.5018}, {"lat": 34.7336, "lng": 135.5019}, {"lat": 34.7337, "lng": 135.5021}, {"lat": 34.734, "lng": 135.5024}, {"lat": 34.7343, "lng": 135.5026}, {"lat": 34.7346, "lng": 135.5024}, {"lat": 34.7347, "lng": 135.5024}, {"lat": 34.7346, "lng": 135.5024}, {"lat": 34.7345, "lng": 135.5025}, {"lat": 34.7343, "lng": 135.5027}, {"lat": 34.7343, "lng": 135.5029}, {"lat": 34.7342, "lng": 135.5031}, {"lat": 34.7344, "lng": 135.5035}, {"lat": 34.7344, "lng": 135.5035}, {"lat": 34.7344, "lng": 135.5035}] (1 row) で登場する、最初の{"lat": 34.7325, "lng": 135.4986}と、最後の {"lat": 34.7344, "lng": 135.5035}だけを取り出すSQL文を書いて下さい
WITH places AS (
SELECT jsonb_array_elements(place) AS place_elements
FROM records
WHERE id = 2
)
SELECT
(SELECT place_elements FROM places LIMIT 1) AS first_place,
(SELECT place_elements FROM places ORDER BY jsonb_array_length(place) DESC LIMIT 1) AS last_place
FROM records
WHERE id = 2;

select user_id from records; で出てくる異なるuser_id(整数)をピックアップするSQL文を教えて下さい
SELECT DISTINCT user_id FROM records;

ありがとうございます。では、これを降順に並べる方法も教えてください
SELECT DISTINCT user_id FROM records ORDER BY user_id DESC;
昇順ならどうなりますか
SELECT DISTINCT user_id FROM records ORDER BY user_id ASC;