Digits = load_digits()
XTrain, XTest, YTrain, YTest = train_test_split(Digits.data, Digits.target,random_state=0)
seaborn.set()
# set up the figure
fig = matplotlib.pyplot.figure(figsize=(6, 6))
fig.subplots_adjust(left=0, right=1, bottom=0, top=1, hspace=0.05, wspace=0.05)
# plot images of Digits
for i in range(64):
ax = fig.add_subplot(8, 8, i + 1, xticks=[], yticks=[])
ax.imshow(Digits.images[i], cmap=matplotlib.pyplot.cm.binary, interpolation='nearest')
# label the image with the value it represents
ax.text(0, 7, str(Digits.target[i]))
model = RandomForestClassifier(n_estimators=1000)
model.fit(XTrain, YTrain)
YPredict = model.predict(XTest)
print(metrics.classification_report(YPredict, YTest))
mat = confusion_matrix(YTest, YPredict)
seaborn.heatmap(mat.T, square=True, annot=True, fmt='d', cbar=False)
matplotlib.pyplot.xlabel('true label')
matplotlib.pyplot.ylabel('predicted label');
#show plots
matplotlib.pyplot.show()
history = model.fit(
np.array(X_train), np.array(y_train),
batch_size = 100,
epochs = 50, verbose=2, validation_data = (X_test, y_test),
)
# %% [markdown]
# ### CNN Predict
# %%
score = model.evaluate(X_test, y_test, verbose = 0)
print('Test loss:', score[0])
print('Test accuracy:', score[1])
pred = model.predict(X_test)
pred = np.argmax(pred, axis = 1)
label = (y_test)
print(pred)
print(label)
# %% [markdown]
# ## Color Clustering with K-Means
# ### Image compression using K-Means can be easy.
# %%
Your_Image_Path = '../Data/vex.PNG'
from tensorflow.keras.preprocessing.image import load_img,img_to_array
img = load_img(Your_Image_Path)
print(img.mode) # RGB color channels
