xeze

import numpy as np
import pandas as pd

import matplotlib.pyplot as plt

from sklearn import datasets
from tensorflow.keras.utils import to_categorical
#Открыть набор данных в переменную
df = pd.read_csv('vehicle.dat')
print(df.columns)
X = df.drop([' Class'], axis = 1)
X.head()
#Y = to_categorical(df[' Class'])
Y = df[' Class']
Y = pd.get_dummies(Y, columns=[' Class'])
Y.head()
#print(Y)
from scipy.spatial.distance import pdist, squareform
import seaborn as sns
from sklearn.preprocessing import StandardScaler, MinMaxScaler
data_scaled = MinMaxScaler().fit_transform(X.to_numpy())
import tensorflow as tf
def plot_the_loss_curve(epochs,mse):
  """График лосс-значений/эпохи"""

  plt.figure()
  plt.xlabel("Epoch")
  plt.ylabel("Mean Squared Error")

  plt.plot(epochs, mse, label="Loss")
  plt.legend()
  plt.ylim([mse.min()*0.97, mse.max()])
  plt.show()

print("Defined the plot_the_model and plot_the_loss_curve functions.")
def build_model(my_learning_rate, X, Y):
  """Создание классификатора"""
  model = tf.keras.models.Sequential()

  model.add(tf.keras.layers.Dense(units=10,
                                  input_shape=(X.shape[1],),activation='tanh'))
  model.add(tf.keras.layers.Dense(10,activation='tanh'))

  # Выходной слой с числом нейронов по числу классов
  model.add(tf.keras.layers.Dense(Y.shape[1], activation = 'softmax'))

  # Лосс-функция уже другая
  model.compile(optimizer=tf.optimizers.Adam(lr=my_learning_rate),
              loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),
              metrics=['accuracy'])

  return model


def train_model(model, feature, label, epochs, batch_size):
  """Обучение модели"""

  history = model.fit(x=feature,
                      y=label,
                      batch_size=batch_size,
                      epochs=epochs)

  trained_weight = model.get_weights()[0]
  trained_bias = model.get_weights()[1]

  epochs = history.epoch

  hist = pd.DataFrame(history.history)

  acc = hist["accuracy"]

  return trained_weight, trained_bias, epochs, acc

print("Defined create_model and train_model")
my_features = data_scaled
my_labels = Y
learning_rate=0.05
epochs=100
my_batch_size= 10

my_model = build_model(learning_rate, X, Y)
trained_weight, trained_bias, epochs, acc = train_model(my_model, my_features,
                                                        my_labels, epochs,
                                                        my_batch_size)
from sklearn.datasets import fetch_openml
from sklearn.preprocessing import StandardScaler
dataset = fetch_openml('har')
X = StandardScaler().fit_transform(dataset.data)
Y = dataset.target
Y_new = []
for i in range(len(Y)):
  Y_new.append(int(Y[i]))