rdisipio

plt.plot(history1.history['binary_accuracy'], color='red')
plt.plot(history1.history['val_binary_accuracy'], color='red', linestyle='dashed')
plt.plot(history2.history['binary_accuracy'], color='blue')
plt.plot(history2.history['val_binary_accuracy'], color='blue', linestyle='dashed')
plt.ylabel('accuracy')
plt.xlabel('epoch')
plt.legend(['train 1', 'val 1', 'train 2', 'val 2'], loc='lower right')
plt.ylim([0.8, 1])
plt.show()

rdisipio

embed_dim = 16

batch_size = 32
n_epochs = 20

import tensorflow as tf
from tensorflow.keras.layers import Conv1D, Flatten, Dense, Dropout

model2 = tf.keras.models.Sequential([
       QuantizedFeaturesEmbedding(n_features, n_bins, embed_dim),

rdisipio

from tensorflow.keras import layers
from tensorflow.keras import initializers

class QuantizedFeaturesEmbedding(layers.Layer):
  def __init__(self,
               n_features,
               n_bins,
               embed_dim,
               **kwargs):
    super(QuantizedFeaturesEmbedding, self).__init__(**kwargs)

rdisipio

from sklearn.preprocessing import KBinsDiscretizer

n_bins = 10 # the more, the merrier
scaled_feature_names = [f"q_{x}" for x in feature_names]

qt = KBinsDiscretizer(n_bins=n_bins, encode='ordinal', strategy='quantile')
X_qt = qt.fit_transform(X).astype(np.int32)  # nb: bin number must be an integer

rdisipio

plt.plot(history1.history['loss'])
plt.plot(history1.history['val_loss'])
plt.title('model loss')
plt.ylabel('loss')
plt.xlabel('epoch')
plt.legend(['train', 'validation'], loc='upper left')
plt.ylim([0., 1.])
plt.show()

rdisipio

model1 = tf.keras.Sequential([
    tf.keras.layers.Dense(128, activation='relu'),
    tf.keras.layers.Dense(2, activation='softmax') # binary classification
])

loss = tf.keras.losses.BinaryCrossentropy(from_logits=False)
metrics = [tf.keras.metrics.BinaryAccuracy()]

model1.compile(optimizer='adam',
              loss=loss,

rdisipio

import numpy as np
import tensorflow as tf

from sklearn import preprocessing

X_scaled = preprocessing.StandardScaler().fit_transform(X)
y_onehot = preprocessing.OneHotEncoder(sparse=False).fit_transform(y.reshape(-1, 1))

(
    X_train, 

rdisipio

from dtreeviz.trees import dtreeviz

viz = dtreeviz(clf, X_test, y_test,
                target_name="target",
                feature_names=feature_names,
                class_names=class_names)

viz

rdisipio

from sklearn.tree import DecisionTreeClassifier

from sklearn.model_selection import train_test_split # Import train_test_split function
from sklearn import metrics

(
    X_train, 
    X_test, 
    y_train, 
    y_test

rdisipio

import matplotlib.pyplot as plt

plot = plt.scatter(X_pca[:,0], X_pca[:,1], c=y)
plt.legend(handles=plot.legend_elements()[0], labels=class_names)