def p3():
# load data
data = pd.read_csv("finances.csv")
# quarterly basis
dates = pd.period_range("1978-09", periods=84, freq="Q")
data.index = dates
# plot data
data.drop([], axis=1).plot(linewidth=1)
plt.xlabel("Date")
plt.ylabel("Earnings and Expenses")
plt.title("Earnings and Expenses per Quarter")
plt.show()
def p1():
# load data
data = pd.read_csv("DJIA.csv")
# set index as datetime
date_index = pd.to_datetime(data["DATE"], format="%Y-%m-%d")
data.index = date_index
df = data.drop(columns=["DATE"])
# drop empty rows and change to floats
df = df.dropna(subset=['VALUE'])
df = df[(df["VALUE"] != ".")]
df["VALUE"] = df["VALUE"].astype('float')
plt.plot(df["VALUE"], lw=0.5)
plt.show()
def p6():
# load data
data = pd.read_csv("DJIA.csv")
# set index as datetime
date_index = pd.to_datetime(data["DATE"], format="%Y-%m-%d")
data.index = date_index
df = data.drop(columns=["DATE"])
# drop empty rows and change to floats
df = df.dropna(subset=['VALUE'])
df = df[(df["VALUE"] != ".")]
df["VALUE"] = df["VALUE"].astype('float')
# plot data
windows = [30, 120, 365]
plt.figure(figsize=(10, 8))
plt.plot(df, alpha=0.5, label='actual')
for w in windows:
plt.plot(df.rolling(window=w).max(), alpha=0.5, label=f'window = {w}')
plt.title('Rolling maximums')
plt.legend()
plt.show()
def p5():
# load data
data = pd.read_csv("DJIA.csv")
# set index as datetime
date_index = pd.to_datetime(data["DATE"], format="%Y-%m-%d")
data.index = date_index
df = data.drop(columns=["DATE"])
# drop empty rows and change to floats
df = df.dropna(subset=['VALUE'])
df = df[(df["VALUE"] != ".")]
df["VALUE"] = df["VALUE"].astype('float')
# find difference for each day and order by value
diff = df - df.shift(1)
s_g = diff.sort_values("VALUE", ascending=False).index[0]
s_l = diff.sort_values("VALUE", ascending=True).index[0]
# find difference for each month and order by value
m_data = df.resample('M').first()
m_diff = (m_data - m_data.shift(1)).dropna()
m_g = m_diff.sort_values("VALUE", ascending=False).index[0]
m_l = m_diff.sort_values("VALUE", ascending=True).index[0]
print("The single day with the largest gain was " + str(s_g) +
". The single day with the largest loss was " + str(s_l) +
". The month with the largest gain was " + str(m_g) +
". The month the the largest loss was " + str(m_l) + ".")
#%%Case 3a: Unbalanced datasets for classification purpose. Following the case 1, here is the equivalent solution: from sklearn.model_selection import train_test_split X_train, X_test, y_train, y_test = train_test_split(X, y, stratify=y, test_size=0.3) #%%% import pandas as pd from sklearn.model_selection import train_test_split datafile_name = 'path_to_data_file' data = pd.read_csv(datafile_name) data=df #target_attribute = data['column_name'] target_attribute = data['mpg'] data = data.drop(columns = ['mpg'], axis = 1) X_train, X_test, y_train, y_test = train_test_split(data, target_attribute, test_size=0.2) X_train X_test y_train y_test df.applymap(lambda x: len(str(x))) pd.applymap? pd.applymaplen(X_train,X_test) #%%%To split into more than two classes such as train, test, and validation, one can do: probs = np.random.rand(len(df)) training_mask = probs < 0.7 test_mask = (probs>=0.7) & (probs < 0.85) validation_mask = probs >= 0.85 probs
import numpy as np
import keras
column_names = [
"subject#", "age", "sex", "test_time", "motor_UPDRS", "total_UPDRS",
"Jitter(%)", "Jitter(Abs)", "Jitter:RAP", "Jitter:PPQ5", "Jitter:DDP",
"Shimmer", "Shimmer(dB)", "Shimmer:APQ3", "Shimmer:APQ5", "Shimmer:APQ11",
"Shimmer:DDA", "NHR", "HNR", "RPDE", "DFA", "PPE"
]
data = pd.read_csv("parkinsons.csv")
df = pd.DataFrame(data, columns=column_names)
data = df.drop('subject#', axis=1)
data1 = data.drop('age', axis=1)
data2 = data1.drop('sex', axis=1)
data3 = data.drop('test_time', axis=1)
X = data3[[
"motor_UPDRS", "Jitter(%)", "Jitter(Abs)", "Jitter:RAP", "Jitter:PPQ5",
"Jitter:DDP", "Shimmer", "Shimmer(dB)", "Shimmer:APQ3", "Shimmer:APQ5",
"Shimmer:APQ11", "Shimmer:DDA", "NHR", "HNR", "RPDE", "DFA", "PPE"
]]
y = data3[["total_UPDRS"]]
x_train, x_test, y_train, y_test = train_test_split(X, y, test_size=0.2)
def get_model():