A set of custom python modules for friendly workflow on pandas
-
Updated
Feb 20, 2024 - Python
A set of custom python modules for friendly workflow on pandas
creation of an interface and algorithm for forecasting demand for 14 days for goods of own production
Effort here is to identify important features in relation to our target variable using multiple Correlation methods based on data type. Feature selection is important for ML models to avoid 'curse of dimensionality' but for this dataset we will be using it build our intution that benefits our later EDA effort
Exploratory Data Analysis of MIMIC-II Dataset and extracted machine learning features
ENG/RU Exploratory Data Analysis "Moscow's Taste: Venue Analysis and Recommendations for Investors" in the Jupyter notebook. The analysis uses libraries such as pandas, seaborn, matplotlib, and phik.
DA рынка HoReCa Москвы: предобработка данных, проверка гипотез и рекомендации для открытия нового заведения (pandas, визуализации, phi_k).
Оценка риска попадания в ДТП по маршруту движения каршеринга
Yandex Practicum Data Science project
Creating a ML-model to forecast the presence of a heart disease
Прогнозирование покупательской активности клиентов
Модель прогнозирования выживаемости стартапов с F1 0.997 на основе DecisionTree, оптимизированного через Optuna. Использованы Python, Scikit-learn, Optuna.
Building a model for predicting ad clicks
Модель прогнозирования риска ДТП для каршеринга с F1 0.674 на основе нейронной сети. Использованы Python, PyTorch, CatBoost, PostgreSQL, Streamlit.
Нейросетевая модель для предсказания температуры звёзд с RMSE 4263 на основе PyTorch. Использованы Python, Scikit-learn, Pandas, корреляция Phik.
Add a description, image, and links to the phik topic page so that developers can more easily learn about it.
To associate your repository with the phik topic, visit your repo's landing page and select "manage topics."