from google.

colab import drive

drive.mount('/content/drive')
import cv2
import matplotlib.pyplot as plt
path="/content/drive/MyDrive/NUUME/人工智慧/Flower/training"
dirs = {"4_tulip": 4, "3_sunflower": 3, "2_rose": 2, "1_dandelion": 1, "0_daisy": 0}
img = cv2.imread("/content/drive/MyDrive/NUUME/人工智
慧/Flower/training/0_daisy/10140303196_b88d3d6cec.jpg")#
#img = cv2.imread("/content/drive/MyDrive/Colab
Notebooks/data/Mid_EX/train/flower_data/test/1128626197_3f52424215_n.jpg")
plt.imshow(img, cmap="gray")
resized_img = cv2.resize(img, (144,144), interpolation = cv2.INTER_AREA)
plt.imshow(resized_img, cmap="gray")
#-----------------------------------------------------------------------------------------------------------
import glob
from tqdm import tqdm
import os
y_ = []
data_x = []
classN = next(os.walk(path))
dim = (128, 128)
#-----------------------------------------------------------------------------------------------------------
for i in range(len(classN[1])):
class_path = path + '/' + classN[1][i] + '/*.*'
print(class_path)
img_path = glob.glob(class_path)
print(img_path)

for j in tqdm(range(len(img_path))):
y_.append(dirs[classN[1][i]])
img = cv2.imread(img_path[j], 1)
resized_img = cv2.resize(img, dim, interpolation = cv2.INTER_AREA)
data_x.append(resized_img)
#-----------------------------------------------------------------------------------------------------------
import pandas as pd
import numpy as np
data_DF = pd.get_dummies(y_)
y_one_hot=np.array(data_DF, dtype='float32')
data_x = np.array(data_x, dtype='float32')
x_= data_x/255
print(x_.shape)
print(y_one_hot.shape)
# 使用 sklearn 的 train_test_split 之前，先洗牌圖像和標籤
from sklearn.utils import shuffle
data_x, y_one_hot = shuffle(data_x, y_one_hot, random_state=42)
from sklearn.model_selection import train_test_split
#-------------- training set, testing set 分割-----------------------------
x_train, x_test, y_train, y_test = train_test_split(data_x, y_one_hot, test_size=0.10,
random_state=42)
#---------------------check data----------------------------------
print("training set data dimension")
print(x_train.shape)
print(y_train.shape)
print("-----------")
print("training set: %i" % len(x_train))
print("testing set: %i" % len(x_test))
print(x_test.shape)
print(y_test.shape)
#-----------------------------------------------------------------------------------------------------------
import cv2
from google.colab.patches import cv2_imshow
no_pitc=3100

#gray cv2. evtColor (x train[no_pitc. 1.

imgg=x_train[no_pitc, :, :, :]
cv2_imshow(imgg)
print(y_train[no_pitc])
#-----------------------------------------------------------------------------------------------------------
import cv2
import matplotlib.pyplot as plt
import glob
from tqdm import tqdm
import os
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from sklearn.model_selection import train_test_split
import tensorflow as tf
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense, Flatten
from tensorflow.keras.optimizers import Adam
from tensorflow.keras.layers import Dense, Flatten, Conv2D, MaxPooling2D
#-----------------------------------------------------------------------------------------------------------
# 定義超參數
batch_size = 64
learning_rate = 0.0001
epochs = 40

# 正規化像素值
x_train_normalized = x_train / 255.0
x_test_normalized = x_test / 255.0

# 定義模型
model = Sequential([
Conv2D(256, (3, 3), activation='relu', input_shape=(128, 128, 3)),
MaxPooling2D((2, 2)),
Conv2D(128, (3, 3), activation='relu'),
MaxPooling2D((2, 2)),
Conv2D(64, (3, 3), activation='relu'),
MaxPooling2D((2, 2)),
Conv2D(32, (3, 3), activation='relu'),
MaxPooling2D((2, 2)),
Flatten(),
Dense(32, activation='relu'),
Dense(5, activation='softmax') # 5 classes for 5 types of flowers
])
# 編譯模型
model.compile(optimizer=Adam(learning_rate=learning_rate),
loss='categorical_crossentropy',
metrics=['accuracy'])

# 訓練模型
history = model.fit(x_train_normalized, y_train,
epochs=epochs,
 batch_size=batch_size,
validation_data=(x_test_normalized, y_test))
model.save('model_CNN_0607T1.keras')
# 繪製訓練和驗證的損失及準確率曲線
plt.figure(figsize=(12, 4))

plt.subplot(1, 2, 1)
plt.plot(history.history['loss'], label='Training Loss')
plt.plot(history.history['val_loss'], label='Validation Loss')
plt.title('Loss')
plt.xlabel('Epoch')
plt.ylabel('Loss')
plt.legend()

plt.subplot(1, 2, 2)
plt.plot(history.history['accuracy'], label='Training Accuracy')
plt.plot(history.history['val_accuracy'], label='Validation Accuracy')
plt.title('Accuracy')
plt.xlabel('Epoch')
plt.ylabel('Accuracy')
plt.legend()

plt.show()
#-----------------------------------------------------------------------------------------------------------
# Plot training and validation accuracy and loss
acc = history.history['accuracy']
val_acc = history.history['val_accuracy']
loss = history.history['loss']
val_loss = history.history['val_loss']
epochs_range = range(25)
# Predict on the test set
predictions = model.predict(x_test)
predicted_classes = np.argmax(predictions, axis=1)
true_classes = np.argmax(y_test, axis=1)
#-----------------------------------------------------------------------------------------------------------
# Save the predictions to a CSV file
results_df = pd.DataFrame({
'True Class': true_classes,
 'Predicted Class': predicted_classes
})
results_df.to_csv('/content/drive/MyDrive/NUUME/人工智慧/Flower/results_0607-
1.csv', index=False)

print("Predictions saved to results.csv")

#-----------------------------------------------------------------------------------------------------------
from tensorflow.keras.models import load_model
import os
import cv2
import numpy as np
import pandas as pd

# 載入訓練好的模型
model = load_model('model_CNN_0607T1.keras')

# 讀取測試資料
csv_path = '/content/drive/MyDrive/NUUME/人工智
慧/Flower/submission_answer.csv'
if not os.path.exists(csv_path):
raise FileNotFoundError(f"CSV file not found: {csv_path}")

test_data = pd.read_csv(csv_path)

# 確認資料格式
if 'id' not in test_data.columns:
raise ValueError("CSV file must contain 'id' column")

test_images = test_data['id'].astype(str)
path = "/content/drive/MyDrive/NUUME/人工智慧/Flower/test"

# 初始化列表以存儲預測結果
predictions = []

# 迭代每個圖像路徑
for image_name in test_images:
image_name_with_extension = image_name + ".jpg"
# 構建完整圖像路徑
 image_path = os.path.join(path, image_name_with_extension)
# 檢查圖像是否存在
if not os.path.exists(image_path):
print(f"Image not found: {image_path}")
continue
# 讀取圖像
image = cv2.imread(image_path)
# 如果圖像讀取成功
if image is not None:
print(f"Reading image: {image_path}")
# 對圖像進行大小調整
resized_img = cv2.resize(image, (128, 128), interpolation=cv2.INTER_AREA)
# 正規化圖像像素值
normalized_img = resized_img / 255.0
# 將圖像轉換為模型所需的格式：(1, height, width, channels)
input_image = np.expand_dims(normalized_img, axis=0)
# 使用模型進行預測
prediction = model.predict(input_image)
# 取得預測結果中概率最高的類別索引
predicted_class_index = np.argmax(prediction)
# 將預測結果添加到列表中
predictions.append(predicted_class_index)
print(f"Prediction for image {image_name}: {predicted_class_index}")
else:
print(f"Failed to read image: {image_path}")

# 將預測結果列表轉換為 NumPy 陣列
predictions = np.array(predictions)

print("Predictions:", predictions)
#-----------------------------------------------------------------------------------------------------------
true_name = test_data.iloc[:, 1].astype(float)
#print(true_name)
k=0
for i in range(len(true_name)):
if true_name[i] == predictions[i]:
k=k+1
#print("correct")
 #else:
#print("wrong")
acc = (k/len(true_name))*100
print("Accuracy: {}%".format(acc))