Data science, machine learning, and visualization projects showcasing Python's power for turning complex data into actionable insights.
Predictive modeling, regression analysis, and classification algorithms to identify patterns and make data-driven predictions.
Creating insightful visualizations with matplotlib, seaborn, and other libraries to communicate complex data relationships effectively.
Automated financial modeling, break-even analysis, and report generation for business decision-making and planning.
Python-based data processing and transformation for AG Grid column customization, enabling dynamic UI configuration and data visualization.
Over the past 5+ years, I've used Python for data analysis, machine learning, and visualization projects across finance, marketing, and product analytics. My approach combines statistical rigor with practical business application, focusing on creating actionable insights from complex datasets.
Pandas, NumPy, Scikit-learn, PyTorch, Matplotlib, Seaborn, SciPy, TensorFlow
Leveraging Python's machine learning libraries to build predictive models and extract insights
A predictive model that analyzes portfolio characteristics to identify accounts that may require rebalancing or special attention. This model was implemented in PowerQuery for portfolio management teams to proactively address client needs.
# Training logistic regression model
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LogisticRegression
# Select features from dataset
features = ['AccountGroupValueDollars',
'CashBufferPercent', 'ClassDrift',
'CategoryDrift', 'PositionDrift',
'TaxType', 'AccountType',
'ModelPositionsNotHeld', 'Equivalencies']
X = dataset[features]
y = dataset[['Columns']]
# Split data into train and test dataset
X_train, X_test, y_train, y_test = train_test_split(X, y)
# Train logistic regression model
log = LogisticRegression()
log.fit(X_train, y_train)
# Testing the algorithm
y_pred_test = log.predict(X_test)
y_prob_test = log.predict_proba(X_test)
# Predict for all inputs
y_pred = log.predict(X)
y_prob = log.predict_proba(X)
# Output results with probabilities
dataset2 = dataset[features + ['Columns']]
dataset2['predictions'] = y_pred
dataset2['probability'] = y_prob[:,1]Creating visual representations to uncover patterns in complex datasets
A heatmap visualization revealing relationships between proposal sections and client engagement metrics.
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
# Calculate the correlation matrix
correlation_matrix = dataset.corr(method='pearson')
# Create a heatmap
plt.figure(figsize=(24, 12))
heatmap = sns.heatmap(correlation_matrix,
annot=True,
cmap="BuPu")
plt.title('Correlation Matrix', fontsize=34)
plt.show()A dendrogram showing hierarchical relationships between investment proposal features and identifying natural content groupings.
import numpy as np import matplotlib.pyplot as plt from scipy.cluster.hierarchy import ward, dendrogram import seaborn as sns # Calculate correlation and distance matrices correlation_matrix = dataset.corr(method='pearson') distance_matrix = 1 - correlation_matrix # Create linkage matrix using Ward's method linkage_matrix = ward(distance_matrix) plt.figure(figsize=(24, 13)) dendrogram(linkage_matrix, labels=correlation_matrix.columns) plt.show()
These visualizations revealed important correlations between proposal sections and client engagement, with "Risk vs Return" and "Hypothetical Growth" sections showing strong client interest.
Historical performance sections had strongest correlation with proposal acceptance.
Risk visualizations clustered separately from growth projections.
Customized sections showed higher engagement than templates.
Automated financial modeling, break-even analysis, and report generation using Python.
This project demonstrates a Python-based system for comprehensive financial analysis for a bakery, covering data preparation, multi-dimensional visualizations (trends, monthly performance), detailed calculations like break-even points, and automated PDF executive report generation.
Leveraging Python for dynamic data processing and UI configuration in advanced data grids.
This project focuses on using Python for backend data processing and transformation to dynamically configure AG Grid columns. It allows for flexible UI based on user roles or data characteristics, enhancing data visualization and interaction within complex enterprise applications.