def get_sensor(n):
""" Get the sensor 8 features, Batch ID, GAS and concentration,
by index
"""
df = load_data()
df_features = get_sensor_features(df, n)
df_gas = df[['Batch ID', 'GAS', 'CONCENTRATION']]
return pd.concat([df_features, df_gas], axis=1)
def get_sensor_by_col_name(n):
"""
Get the sensor 8 features, Batch ID, GAS and concentration,
by column names
"""
df = load_data()
filter_col = [col for col in df if col.startswith(f'S{n}')]
df_sensor = df[filter_col]
df_gas = df[['Batch ID', 'GAS', 'CONCENTRATION']]
return pd.concat([df_sensor, df_gas], axis=1)
def get_sensors_list(list_of_n):
"""
Get the N sensor features (N*8), Batch ID, GAS and concentration,
by index
:arg list of integers
:return pandas dataframe
"""
df = load_data()
sensors_columns = []
for n in list_of_n:
ix = list(sensor_features_column(n))
sensors_columns.extend(ix)
#sensors_columns = np.r_[1:10, 15, 17, 50:100]
df_features_n = df.iloc[:, sensors_columns]
df_gas = df[['Batch ID', 'GAS', 'CONCENTRATION']]
return pd.concat([df_features_n, df_gas], axis=1)
model = seq.load_model(model_name)
ev = Evolution_drift(df_gas, n_batch_training, batch)
_, X_test, _, y_test = ev.split_data()
loss, acc = seq.model_evaluate(X_test, y_test)
model_dict[batch] = {'acc': acc, 'loss': loss}
#Plot results
df_results = pd.DataFrame.from_dict(model_dict).T
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(12, 8))
fig.suptitle(f'Training with first {n_batch_training} batches')
ax1 = df_results.plot(kind='bar', y='acc', ax=ax1)
ax1.set_ylim([0, 1])
ax2 = df_results.plot(kind='bar', y='loss', ax=ax2)
save_figure(fig, f'Step1_NBATCH_{n_batch_training}_acc_loss')
if __name__ == '__main__':
# Check drift relevance with different training sets.
df_gas = load_data()
for col in ['GAS', 'Batch ID']:
df_gas[col] = df_gas[col].astype('category')
for n in range(1, 10 + 1):
evo_train(df_gas, n_batch_training=n)
for n in range(1, 10 + 1):
evolution_drift_with_seq(df_gas, n_batch_training=n)
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from sklearn.preprocessing import StandardScaler
from sklearn.cluster import KMeans
from sklearn.decomposition import PCA
from python.LoadUciData import load_data
from python.StandardFigure import save_figure
from python.Step0_1_DataExploration import get_sensors_array
if __name__ == '__main__':
# Cargo datos
df = load_data()
arr_sensor = get_sensors_array(df, 1, 0)
pca = PCA(n_components=2)
pca.fit(arr_sensor)
arr_transf = pca.transform(arr_sensor)
number_of_clusters = 4
km = KMeans(n_clusters=number_of_clusters)
# Normally people fit the matrix
y_pred = km.fit_predict(arr_transf)
plt.scatter(arr_transf[:, 0], arr_transf[:, 1], c=y_pred)
plt.title("Cluster of sensors")
plt.show()
