驗證損失和驗證准確度曲線隨預訓練 Model 波動

Question

我目前正在研究神經網絡，在嘗試學習 CNN 時遇到問題，我正在嘗試訓練包含有關音樂流派的頻譜圖的數據。 我的數據由 27000 個頻譜圖組成，分為 3 個 class（流派）。 我的數據以 9:1 的比例拆分用於訓練和驗證

誰能幫助我，為什么我的驗證損失/准確性的結果會波動？ 我正在使用 Keras 的 MobileNetV2 並將其與 3 個密集層連接。 這是我的代碼片段：

train_datagen = ImageDataGenerator(
    preprocessing_function=preprocess_input,
    validation_split=0.1)

train_generator = train_datagen.flow_from_dataframe(
    dataframe=traindf,
    directory="...",
    color_mode='rgb',
    x_col="ID",
    y_col="Class",
    subset="training",
    batch_size=32,
    seed=42,
    shuffle=True,
    class_mode="categorical",
    target_size=(64, 64))

valid_generator = train_datagen.flow_from_dataframe(
    dataframe=traindf,
    directory="...",
    color_mode='rgb',
    x_col="ID",
    y_col="Class",
    subset="validation",
    batch_size=32,
    seed=42,
    shuffle=True,
    class_mode="categorical",
    target_size=(64, 64))

base_model = MobileNetV2(weights='imagenet', include_top=False)
x = base_model.output
x = GlobalAveragePooling2D()(x)
x = Dense(1025, activation='relu')(x)
x = Dense(1025, activation='relu')(x)
x = Dense(512, activation='relu')(x)
preds = Dense(3, activation='softmax')(x)
model = Model(inputs=base_model.input, outputs=preds)

model.compile(optimizer='adam', loss='categorical_crossentropy',
                  metrics=['accuracy'])

step_size_train = train_generator.n//train_generator.batch_size
step_size_valid = valid_generator.n//valid_generator.batch_size
history = model.fit_generator(
    generator=train_generator,
    steps_per_epoch=step_size_train,
    validation_data=valid_generator,
    validation_steps=step_size_valid,
    epochs=75)

這些是我的驗證損失和驗證准確度曲線波動太大的圖片

有沒有辦法減少波動或讓它變得更好？ 我在這里有過度擬合或擬合不足的問題嗎？ 我曾嘗試使用 Dropout() ，但它只會讓情況變得更糟。 我需要做什么來解決這個問題？

感謝之前，Aquilla Setiawan Kanadi。

Answer 1

首先，缺少驗證損失和驗證准確性的圖片。

要回答您的問題，以下可能是您的驗證丟失和驗證准確性波動的原因 -

1. 您為 base_model 添加了大約 1.25 倍的權重來構建 model。 (model Trainable Parameters 5115398 - base_model Trainable Parameters 2223872 = 2891526)

節目統計：

import tensorflow as tf
from tensorflow.keras.models import Model
from tensorflow.keras.layers import GlobalAveragePooling2D, Dense
from keras.utils.layer_utils import count_params

class color:
   PURPLE = '\033[95m'
   CYAN = '\033[96m'
   DARKCYAN = '\033[36m'
   BLUE = '\033[94m'
   GREEN = '\033[92m'
   YELLOW = '\033[93m'
   RED = '\033[91m'
   BOLD = '\033[1m'
   UNDERLINE = '\033[4m'
   END = '\033[0m'

base_model = tf.keras.applications.MobileNetV2(weights='imagenet', include_top=False)

#base_model.summary()
trainable_count = count_params(base_model.trainable_weights)
non_trainable_count = count_params(base_model.non_trainable_weights)
print("\n",color.BOLD + '  base_model Statistics !' + color.END)
print("Trainable Parameters :", color.BOLD + str(trainable_count) + color.END)
print("Non Trainable Parameters :", non_trainable_count,"\n")

x = base_model.output
x = GlobalAveragePooling2D()(x)
x = Dense(1025, activation='relu')(x)
x = Dense(1025, activation='relu')(x)
x = Dense(512, activation='relu')(x)
preds = Dense(3, activation='softmax')(x)
model = Model(inputs=base_model.input, outputs=preds)

#model.summary()
trainable_count = count_params(model.trainable_weights)
non_trainable_count = count_params(model.non_trainable_weights)
print(color.BOLD + '    model Statistics !' + color.END)
print("Trainable Parameters :", color.BOLD + str(trainable_count) + color.END)
print("Non Trainable Parameters :", non_trainable_count,"\n")

new_weights_added = count_params(model.trainable_weights) - count_params(base_model.trainable_weights)
print("Additional trainable weights added to the model excluding basel model trainable weights :", color.BOLD + str(new_weights_added) + color.END)

Output -

WARNING:tensorflow:`input_shape` is undefined or non-square, or `rows` is not in [96, 128, 160, 192, 224]. Weights for input shape (224, 224) will be loaded as the default.

   base_model Statistics !
Trainable Parameters : 2223872
Non Trainable Parameters : 34112 

    model Statistics !
Trainable Parameters : 5115398
Non Trainable Parameters : 34112 

Additional trainable weights added to the model excluding basel model trainable weights : 2891526

2. 您正在訓練完整的 model 權重（MobileNetV2 權重和附加層權重）。

您的問題的解決方案是 -

1. 以與 base_model 可訓練參數相比新的可訓練參數最小的方式自定義附加層。 可以添加最大池化層和更少的密集層。

2. 通過base_model.trainable = False凍結基礎 model 並僅訓練您在 MobileNetV2 層之上添加的新層。

或者

解凍基礎模型的頂層（MobileNetV2 層）並將底層設置為不可訓練。 您可以按如下方式進行，我們將 model 凍結到第 100 層，其余層將是可訓練的 -

# Let's take a look to see how many layers are in the base model
print("Number of layers in the base model: ", len(base_model.layers))

# Fine-tune from this layer onwards
fine_tune_at = 100

# Freeze all the layers before the `fine_tune_at` layer
for layer in base_model.layers[:fine_tune_at]:
  layer.trainable =  False

Output -

Number of layers in the base model:  155

3. 通過超參數調整訓練 model。 您可以在此處找到有關超參數調整的更多信息。