Eryk Lewinson erykml

Data Scientist, Author of the Python for Finance Cookbook (published by Packt).

erykml / ts_feature_engineering_2.py

Last active February 17, 2023 15:20

	# for reproducibility
	np.random.seed(42)
	# generate the DataFrame with dates
	range_of_dates = pd.date_range(
	start="2017-01-01",
	end="2020-12-30"
	)
	X = pd.DataFrame(index=range_of_dates)
	# create a sequence of day numbers
	X["day_nr"] = range(len(X))

erykml / ts_feature_engineering_1.py

Created July 28, 2022 20:25

	import numpy as np
	import pandas as pd
	import matplotlib.pyplot as plt
	import seaborn as sns
	from datetime import date
	from sklearn.linear_model import LinearRegression
	from sklearn.preprocessing import FunctionTransformer
	from sklearn.metrics import mean_absolute_error
	from sklego.preprocessing import RepeatingBasisFunction

erykml / data_validation.py

Created July 12, 2022 21:23

	import pandas as pd
	import glob
	import os

	from config import RAW_DIR, PROCESSED_DIR, REPORTS_DIR, AUGMENTED_DIR

	from deepchecks.tabular import Dataset
	from deepchecks.tabular.suites import full_suite

	# prepare output dir for the data validation reports

erykml / data_augmentation.py

Created July 12, 2022 21:23

	import os
	import pandas as pd
	from config import RAW_DIR, PROCESSED_DIR, AUGMENTED_DIR

	from imblearn.over_sampling import RandomOverSampler, SMOTE, ADASYN
	from imblearn.under_sampling import RandomUnderSampler

	RANDOM_STATE = 42

	# define the considered augmentations

erykml / data_preparation.py

Created July 12, 2022 21:22

	import pandas as pd
	from config import RAW_DIR, PROCESSED_DIR
	import os

	from sklearn.preprocessing import RobustScaler
	from sklearn.model_selection import train_test_split

	# load data
	df = pd.read_csv(f"{RAW_DIR}/creditcard.csv")

erykml / config.py

Created July 8, 2022 15:33

	# directories
	RAW_DIR = "data/raw"
	PROCESSED_DIR = "data/processed"
	AUGMENTED_DIR = "data/augmented"
	REPORTS_DIR = "data_validation"

erykml / cbpe_2.py

Created May 11, 2022 09:47

	pred_df["TP"] = np.where(pred_df["class"] == 1, pred_df["prob"], 0)
	pred_df["FP"] = np.where(pred_df["class"] == 1, 1 - pred_df["prob"], 0)
	pred_df["TN"] = np.where(pred_df["class"] == 0, 1- pred_df["prob"], 0)
	pred_df["FN"] = np.where(pred_df["class"] == 0, pred_df["prob"], 0)
	pred_df

erykml / cbpe_1.py

Last active May 11, 2022 09:46

erykml / pyscript_2.html

Created May 1, 2022 19:02

	<html>
	<head>
	<link rel="stylesheet" href="https://pyscript.net/alpha/pyscript.css" />
	<script defer src="https://pyscript.net/alpha/pyscript.js"></script>
	<py-env>
	- numpy
	- matplotlib
	</py-env>
	</head>

erykml / pyscript_1.html

Created May 1, 2022 19:01

	<html>
	<head>
	<link rel="stylesheet" href="https://pyscript.net/alpha/pyscript.css" />
	<script defer src="https://pyscript.net/alpha/pyscript.js"></script>
	</head>
	<body> <py-script> print('Hello, World!') </py-script> </body>
	</html>