Movatterモバイル変換

[0]ホーム
URL:

Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

 Sign up
Appearance settings

Commit810e604

Browse files
committed
add image classification tutorial
1 parent78e6d6d commit810e604

File tree

5 files changed

+188
-1
lines changed

5 files changed

+188
-1
lines changed

‎README.md‎

Lines changed: 2 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -19,9 +19,10 @@ This is a repository of all the tutorials of [The Python Code](https://www.thepy
1919
-###[Machine Learning](https://www.thepythoncode.com/topic/machine-learning)
2020
-###[Natural Language Processing](https://www.thepythoncode.com/topic/nlp)
2121
- [How to Build a Spam Classifier using Keras in Python](https://www.thepythoncode.com/article/build-spam-classifier-keras-python). ([code](machine-learning/nlp/spam-classifier))
22-
- [How to Detect Human Faces in Python using OpenCV](https://www.thepythoncode.com/article/detect-faces-opencv-python). ([code](machine-learning/face_detection))
2322

23+
-[How to Detect Human Faces in Python using OpenCV](https://www.thepythoncode.com/article/detect-faces-opencv-python). ([code](machine-learning/face_detection))
2424
-[Building a Speech Emotion Recognizer using Scikit-learn](https://www.thepythoncode.com/article/building-a-speech-emotion-recognizer-using-sklearn). ([code](machine-learning/speech-emotion-recognition))
25+
-[How to Make an Image Classifier in Python using Keras](https://www.thepythoncode.com/article/image-classification-keras-python). ([code](machine-learning/image-classifier))
2526
-[Top 8 Python Libraries For Data Scientists and Machine Learning Engineers](https://www.thepythoncode.com/article/top-python-libraries-for-data-scientists).
2627

2728
-###[General Python Topics](https://www.thepythoncode.com/topic/general-python-topics)
Lines changed: 14 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,14 @@
1+
#[How to Make an Image Classifier in Python using Keras](https://www.thepythoncode.com/article/image-classification-keras-python)
2+
To run this:
3+
-`pip3 install -r requirements.txt`
4+
- First, you need to train the model using`python train.py`
5+
- Edit the code in`test.py` for you optimal model weights in`results` folder ( currently empty because you need to train first ) and run:
6+
```
7+
python test.py
8+
```
9+
**Output:**
10+
```
11+
10000/10000 [==============================] - 3s 331us/step
12+
Test accuracy: 81.17999999999999 %
13+
frog
14+
```
Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1 @@
1+
keras
Lines changed: 37 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,37 @@
1+
fromtrainimportload_data
2+
fromkeras.modelsimportload_model
3+
importmatplotlib.pyplotasplt
4+
importnumpyasnp
5+
6+
# CIFAR-10 classes
7+
categories= {
8+
0:"airplane",
9+
1:"automobile",
10+
2:"bird",
11+
3:"cat",
12+
4:"deer",
13+
5:"dog",
14+
6:"frog",
15+
7:"horse",
16+
8:"ship",
17+
9:"truck"
18+
}
19+
20+
# load the testing set
21+
(_,_), (X_test,y_test)=load_data()
22+
# load the model with optimal weights
23+
model=load_model("results/cifar10-loss-0.58-acc-0.81.h5")
24+
# evaluation
25+
loss,accuracy=model.evaluate(X_test,y_test)
26+
print("Test accuracy:",accuracy*100,"%")
27+
28+
# get prediction for this image
29+
sample_image=X_test[7500]
30+
prediction=np.argmax(model.predict(sample_image.reshape(-1,*sample_image.shape))[0])
31+
print(categories[prediction])
32+
33+
# show the first image
34+
plt.axis('off')
35+
plt.imshow(sample_image)
36+
plt.savefig("frog.png")
37+
plt.show()
Lines changed: 134 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,134 @@
1+
fromkeras.datasetsimportcifar10# importing the dataset from keras
2+
fromkeras.modelsimportSequential
3+
fromkeras.layersimportDense,Dropout,Activation,Flatten
4+
fromkeras.layersimportConv2D,MaxPooling2D
5+
fromkeras.callbacksimportModelCheckpoint,TensorBoard
6+
fromkeras.utilsimportto_categorical
7+
importos
8+
9+
# hyper-parameters
10+
batch_size=64
11+
# 10 categories of images (CIFAR-10)
12+
num_classes=10
13+
# number of training epochs
14+
epochs=30
15+
16+
defcreate_model(input_shape):
17+
"""
18+
Constructs the model:
19+
- 32 Convolutional (3x3)
20+
- Relu
21+
- 32 Convolutional (3x3)
22+
- Relu
23+
- Max pooling (2x2)
24+
- Dropout
25+
26+
- 64 Convolutional (3x3)
27+
- Relu
28+
- 64 Convolutional (3x3)
29+
- Relu
30+
- Max pooling (2x2)
31+
- Dropout
32+
33+
- 128 Convolutional (3x3)
34+
- Relu
35+
- 128 Convolutional (3x3)
36+
- Relu
37+
- Max pooling (2x2)
38+
- Dropout
39+
40+
- Flatten (To make a 1D vector out of convolutional layers)
41+
- 1024 Fully connected units
42+
- Relu
43+
- Dropout
44+
- 10 Fully connected units (each corresponds to a label category (cat, dog, etc.))
45+
"""
46+
47+
# building the model
48+
model=Sequential()
49+
50+
model.add(Conv2D(filters=32,kernel_size=(3,3),padding="same",input_shape=input_shape))
51+
model.add(Activation("relu"))
52+
model.add(Conv2D(filters=32,kernel_size=(3,3),padding="same"))
53+
model.add(Activation("relu"))
54+
model.add(MaxPooling2D(pool_size=(2,2)))
55+
model.add(Dropout(0.25))
56+
57+
model.add(Conv2D(filters=64,kernel_size=(3,3),padding="same"))
58+
model.add(Activation("relu"))
59+
model.add(Conv2D(filters=64,kernel_size=(3,3),padding="same"))
60+
model.add(Activation("relu"))
61+
model.add(MaxPooling2D(pool_size=(2,2)))
62+
model.add(Dropout(0.25))
63+
64+
model.add(Conv2D(filters=128,kernel_size=(3,3),padding="same"))
65+
model.add(Activation("relu"))
66+
model.add(Conv2D(filters=128,kernel_size=(3,3),padding="same"))
67+
model.add(Activation("relu"))
68+
model.add(MaxPooling2D(pool_size=(2,2)))
69+
model.add(Dropout(0.25))
70+
71+
# flattening the convolutions
72+
model.add(Flatten())
73+
# fully-connected layers
74+
model.add(Dense(1024))
75+
model.add(Activation("relu"))
76+
model.add(Dropout(0.5))
77+
model.add(Dense(num_classes,activation="softmax"))
78+
79+
# print the summary of the model architecture
80+
model.summary()
81+
82+
# training the model using rmsprop optimizer
83+
model.compile(loss="categorical_crossentropy",optimizer="adam",metrics=["accuracy"])
84+
returnmodel
85+
86+
87+
defload_data():
88+
"""
89+
This function loads CIFAR-10 dataset, normalized, and labels one-hot encoded
90+
"""
91+
# loading the CIFAR-10 dataset, splitted between train and test sets
92+
(X_train,y_train), (X_test,y_test)=cifar10.load_data()
93+
print("Training samples:",X_train.shape[0])
94+
print("Testing samples:",X_test.shape[0])
95+
print(f"Images shape:{X_train.shape[1:]}")
96+
97+
# converting image labels to binary class matrices
98+
y_train=to_categorical(y_train,num_classes)
99+
y_test=to_categorical(y_test,num_classes)
100+
101+
# convert to floats instead of int, so we can divide by 255
102+
X_train=X_train.astype("float32")
103+
X_test=X_test.astype("float32")
104+
X_train/=255
105+
X_test/=255
106+
107+
return (X_train,y_train), (X_test,y_test)
108+
109+
110+
if__name__=="__main__":
111+
112+
# load the data
113+
(X_train,y_train), (X_test,y_test)=load_data()
114+
115+
# constructs the model
116+
model=create_model(input_shape=X_train.shape[1:])
117+
118+
# some nice callbacks
119+
tensorboard=TensorBoard(log_dir="logs/cifar10-model-v1")
120+
checkpoint=ModelCheckpoint("results/cifar10-loss-{val_loss:.2f}-acc-{val_acc:.2f}.h5",
121+
save_best_only=True,
122+
verbose=1)
123+
124+
# make sure results folder exist
125+
ifnotos.path.isdir("results"):
126+
os.mkdir("results")
127+
128+
# train
129+
model.fit(X_train,y_train,
130+
batch_size=batch_size,
131+
epochs=epochs,
132+
validation_data=(X_test,y_test),
133+
callbacks=[tensorboard,checkpoint],
134+
shuffle=True)

0 commit comments

Comments
 (0)
[8]ページ先頭
©2009-2025 Movatter.jp