build model

loadAffwildDS.py

@@ -10,12 +10,15 @@ def extractDataFromVideo_(filename, videoName, facesList, labelsList, maxNbrImag
     emotions = ["Neutral", "Angry", "Disgust",
                 "Fear", "Happy", "Sad", "Suprise"]
 
-    # Start capture of a video and reading of a label file
+    # Start capture of a video and reading of a label file. Dont change the lists if file can not be read.
     cap = cv2.VideoCapture(filename)
     if (cap.isOpened() == False):
         print("Error opening video")
 
+    try:
         file = open("data/affwild/labels/"+videoName[:-4]+'.txt', 'r')
+    except FileNotFoundError:
+        return facesList, labelsList
     file.readline()
 
     # Read until video is completed
@@ -37,7 +40,8 @@ def extractDataFromVideo_(filename, videoName, facesList, labelsList, maxNbrImag
                 newFaces = ip.imageProcess(frame, writeEmotion=False)
 
                 # If 2 faces were detected, it means an error was made since there is only single-person videos here.
-                if len(newFaces) == 1:
+                # The second condition means the image is irrelevant (no face on the picture)
+                if len(newFaces) == 1 and line[0] != '-':
                     facesList += newFaces
 
                     emotionNbr = emotionToNumber(emotions[int(line[0])])

utils.py

@@ -41,3 +41,39 @@ def emotionToNumber(emotion):
     emotions = ["Angry", "Disgust", "Fear",
                 "Happy", "Sad", "Suprise", "Neutral"]
     return emotions.index(emotion)
+
+
+def mergeToDatabase(listOfX, listOfY, validation_repart=[0.025, 0.025, 0.025, 0.025]):
+    # This shuffle each X, extract validation data, merge differents X, shuffle again.
+    listOfX_train, listOfY_train = [], []
+    listOfX_test, listOfY_test = [], []
+
+    for X, Y, rate in zip(listOfX, listOfY, validation_repart):
+        N = X.shape[0]
+
+        # Shuffle each X and Y the same way
+        shuffler = np.random.permutation(N)
+        X, Y = X[shuffler], Y[shuffler]
+        # Extract validation data
+        X_train, Y_train = X[:N*(1-rate)], Y[:N*(1-rate)]
+        X_test, Y_test = X[N*(1-rate):], Y[N*(1-rate):]
+        listOfX_train.append(X_train)
+        listOfY_train.append(Y_train)
+        listOfX_test.append(X_test)
+        listOfY_test.append(Y_test)
+    # Merge
+    BigX_train = np.stack(listOfX_train)
+    BigY_train = np.stack(listOfY_train)
+
+    BigX_test = np.stack(listOfX_test)
+    BigY_test = np.stack(listOfY_test)
+    # Shuffle the whole
+    shuffler = np.random.permutation(len(BigX_train))
+    BigX_train = BigX_train[shuffler]
+    BigY_train = BigY_train[shuffler]
+
+    shuffler = np.random.permutation(len(BigX_test))
+    BigX_test = BigX_test[shuffler]
+    BigY_test = BigY_test[shuffler]
+
+    return BigX_train, BigY_train, BigX_test, BigY_test