ia

IA/data/data_emple.csv

@@ -1,6 +1,6 @@
-category,distance,timeOfActivity,denivelePositif,speedAvg,bpmAvg
-walking,2147.0,1716.5,471.5999999999999,1.2740575455079353,114
-walking,1013.0,933.24,393.2000000000004,1.2849015317286667,122
-walking,225.0,228.49,346.6000000000006,1.1007253886010366,99
-cycling,20525.4,6770.675,597.8000000000001,3.5130617816091947,143
+category,distance,timeOfActivity,denivelePositif,speedAvg,bpmAvg,startTime
+walking,2147.0,1716.5,471.5999999999999,1.2740575455079353,114,1234
+walking,1013.0,933.24,393.2000000000004,1.2849015317286667,122,1245
+walking,225.0,228.49,346.6000000000006,1.1007253886010366,99,1270
+walking,20525.4,6770.675,597.8000000000001,3.5130617816091947,143,1300
 

IA/test.py

@@ -0,0 +1,57 @@
+import pandas as pd 
+import numpy as np
+from sklearn.linear_model import LinearRegression
+from sklearn.preprocessing import LabelEncoder
+import pickle
+import sklearn_json as skljson
+
+
+
+
+def generateJson(model:LinearRegression):
+    listCoef = []
+    listIntercept = []
+    for i in range(0,len(model.coef_)):
+        listCoef.append(model.coef_[i][0])
+        listIntercept.append(model.intercept_[i])
+    json = {"coef":listCoef,"intercept":listIntercept}
+    return json
+
+
+
+
+# Load data from CSV
+df = pd.read_csv("data\\data_emple.csv")
+
+startTime = df.iloc[0:len(df),6].values.reshape(-1,1)
+
+category = df.iloc[0:len(df),0].values
+#print("Category : ",category)
+
+data = pd.DataFrame({
+    "Distance": df.iloc[:, 1].values,
+    "Time": df.iloc[:, 2].values,
+    "Denivele": df.iloc[:, 3].values,
+    "Speed": df.iloc[:, 4].values,
+    "Bpm": df.iloc[:, 5].values
+})
+
+model = LinearRegression()
+
+model.fit(startTime,data)
+
+datePredict = np.array([[1271]])
+prediction = model.predict(datePredict)
+
+print("Prédiction -> ")
+#print(prediction)
+print("Distance : ",float(prediction[0][0]))
+print("Time : ",float(prediction[0][1]))
+print("Denivele : ",float(prediction[0][2]))
+print("Speed : ",float(prediction[0][3]))
+print("BPM : ",float(prediction[0][4]))
+
+
+generateJson(model=model)
+
+

lib/modele/user.dart

@@ -1,5 +1,8 @@
+import 'dart:convert';
+
 import 'package:flutter/material.dart';
 import 'package:smartfit_app_mobile/modele/activity.dart';
+import 'package:smartfit_app_mobile/modele/activity_info/activity_info.dart';
 import 'package:smartfit_app_mobile/modele/manager_selected_activity.dart';
 
 class User extends ChangeNotifier {
@@ -83,4 +86,27 @@ class User extends ChangeNotifier {
     map["durationActiviy"] = selected.getTimeAllActivitySelected();
     return map;
   }
+
+  ActivityInfo predictActivity(DateTime date) {
+    // Appel pour avoir le model
+    String jsonString =
+        '{"coef": [270.63861280635473, 74.69699263779908, 1.9946527172333637, 0.03215810401413792, 0.3256805192289063], "intercept": [-335635.9890148213, -91874.0527070619, -2065.450392327813, -38.79838022998388, -291.590235396687]}';
+    Map<String, dynamic> jsonMap = json.decode(jsonString);
+    // Transformer la date
+    int dateMilli = date.millisecondsSinceEpoch;
+
+    ActivityInfo activityInfo = ActivityInfo();
+    activityInfo.distance =
+        jsonMap["coef"][0] * dateMilli + jsonMap["intercept"][0];
+    activityInfo.timeOfActivity =
+        jsonMap["coef"][1] * dateMilli + jsonMap["intercept"][1];
+    activityInfo.denivelePositif =
+        jsonMap["coef"][2] * dateMilli + jsonMap["intercept"][2];
+    activityInfo.vitesseAvg =
+        jsonMap["coef"][3] * dateMilli + jsonMap["intercept"][3];
+    activityInfo.bpmAvg =
+        jsonMap["coef"][4] * dateMilli + jsonMap["intercept"][4];
+
+    return activityInfo;
+  }
 }