problem overfitting model VGG16 small dataset - python-3.x

Question

Screenshot of the problem
i have two classes, each of them contains an equal number of pictures
Train
360 train pictures for classe one
360 train pictures for classe two
Test
90 test pictures classe one
90 test pictures classe two
my code
def load_split(basePath, csvPath):
data = []
labels = []
rows = open(csvPath).read().strip().split("\n")[1:]
random.shuffle(rows)
for (i, row) in enumerate(rows):
if i > 0:
print("[INFO] processed {} total images".format(i))
(label, imagePath) = row.strip().split(",")[-2:]
imagePath = os.path.sep.join([basePath, imagePath])
image = io.imread(imagePath)
image = transform.resize(image, (224, 224))
image = exposure.equalize_adapthist(image, clip_limit=0.1)
data.append(image)
labels.append(int(label))
data = np.array(data)
labels = np.array(labels)
return (data, labels)
ap = argparse.ArgumentParser()
ap.add_argument("-d", "--dataset", required=True,
help="path to input GTSRB")
ap.add_argument("-m", "--model", required=True,
help="path to output model")
ap.add_argument("-p", "--plot", type=str, default="plot.png",
help="path to training history plot")
args = vars(ap.parse_args())
NUM_EPOCHS = 30
INIT_LR = 1e-4
BS = 64
labelNames = open("signnames.csv").read().strip().split("\n")[1:]
labelNames = [l.split(",")[1] for l in labelNames]
trainPath = os.path.sep.join([args["dataset"], "Train.csv"])
testPath = os.path.sep.join([args["dataset"], "Test.csv"])
print("[INFO] loading training and testing data...")
(trainX, trainY) = load_split(args["dataset"], trainPath)
(testX, testY) = load_split(args["dataset"], testPath)
trainX = (trainX-np.mean(trainX))/np.std(trainX)
testX = (testX-np.mean(testX))/np.std(testX)
numLabels = len(np.unique(trainY))
trainY = to_categorical(trainY, numLabels)
testY = to_categorical(testY, numLabels)
classTotals = trainY.sum(axis=0)
classWeight = classTotals.max() / classTotals
aug = ImageDataGenerator(
rotation_range=10,
zoom_range=0.15,
width_shift_range=0.1,
height_shift_range=0.1,
shear_range=0.15,
horizontal_flip=False,
vertical_flip=False,
fill_mode="nearest")
print("[INFO] compiling model...")
opt = Adam(lr=INIT_LR)
model = TrafficSignNet.build(width=224, height=224, depth=3,
classes=numLabels)
model.compile(loss="categorical_crossentropy", optimizer=opt,
metrics=["accuracy"])
print("[INFO] training network...")
H = model.fit_generator(
aug.flow(trainX, trainY, batch_size=BS),
validation_data=(testX, testY),
steps_per_epoch=trainX.shape[0] // BS,
epochs=NUM_EPOCHS,
class_weight=classWeight,
verbose=1)
print("[INFO] evaluating network...")
predictions = model.predict(testX, batch_size=BS)
print(classification_report(testY.argmax(axis=1),
predictions.argmax(axis=1), target_names=labelNames))
print("[INFO] serializing network to '{}'...".format(args["model"]))
model_json = model.to_json()
with open("model.json", "w") as json_file:
json_file.write(model_json)
model.save_weights("model.h5")
print("Saved model to disk")
Enddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddd

Answer 1

I could not get access to your training script for some reason - can check your screenshot only.
I think your problem is not overfitting - your model is actually not learning, since the loss is not reduced while the accuracy keeps at the same level. Please double check you network, learning rate (most important) and preprocessing. Also, you may consider loading the VGG16 pretrained weight or not to load if you have done.
Or just post your code, I can take a look when I get a chance.
Update:
Based on your codes, I found you did not make any changes internally on the VGG16 - which is easy to debug.
Check you training and test set and make sure the classes are evenly distributed
Print out the label (Y test and train), carefully check if they are correct.
Try to standardize the X train and test instead of dividing by 255. x=(x-mean)/std
Try use learning rate as 0.0001 (I found it's generally good for VGG16 based network)
Stay simple at the first time. Don't use the decayed optimization, just try the standard ADAM firstly
Bests,