Local Development Environment

Build and Run Deep Learning Studio Locally

A powerful local setup for developing, training, and deploying deep learning models. Get started in minutes with our streamlined Windows launcher.

  • One-command setup
  • Full-stack workspace
  • Built-in XAI
  • 20+ algorithms
Get Started View on GitHub

Local Setup Guide

Get up and running in minutes with our streamlined setup process

One-Click Local Setup

Clone the repository and run the launcher script. Dependencies are installed automatically, and both backend and frontend services start together.

1
Clone the Repository

Download the DL-Studio project from GitHub

2
Navigate to Project Folder

Move into the DL-Studio directory

3
Run the Launcher

Execute the startup script to begin development

Windows CMD Recommended 
git clone https://github.com/purushothaman-natarajan/DL-Studio.git
cd DL-Studio
run_studio.bat
Copied to clipboard!

Supported Algorithms

Comprehensive collection of traditional ML and deep learning models

Boosting Models

XGBoost LightGBM CatBoost Gradient Boosting

Tree-Based Models

Decision Tree Random Forest Extra Trees

Support Vector Machines

SVM (RBF Kernel) SVM (Linear) SVM (Polynomial)

Linear Models

Linear Regression Ridge Regression Lasso Regression

KNN & Distance-Based

K-Nearest Neighbors Distance Metrics

Neural Networks

MLP (Multi-Layer Perceptron) RNN / LSTM / GRU Transformer

Explainable AI (XAI)

Built-in interpretability techniques for model transparency

SHAP Analysis

SHapley Additive exPlanations for global and local feature importance using KernelExplainer.

  • Global feature impact visualization
  • Local explanation per prediction
  • Weighted feature contributions

LIME Explanations

Local Interpretable Model-agnostic Explanations for understanding individual predictions.

  • Local surrogate models
  • Feature perturbation analysis
  • Per-instance explanations

Sensitivity Analysis

Feature sensitivity curves showing how model output changes with input variations.

  • One-at-a-time perturbation
  • Response surface visualization
  • Critical feature identification

Correlation Matrix

Feature correlation heatmap to understand relationships and multicollinearity.

  • Pearson correlation
  • Feature interaction detection
  • Multicollinearity warning

Residual Analysis

Actual vs Predicted plots to diagnose model performance and error patterns.

  • Prediction error visualization
  • Outlier detection
  • Model fit assessment

Feature Importance

Native feature importance rankings from tree-based and ensemble models.

  • Gini/MDI importance
  • Permutation importance
  • Ranked feature list

Future Roadmap

Planned features and enhancements for upcoming releases

Coming Soon

NLP & Text Models

Support for BERT, GPT, and transformer-based models for text classification, NER, and sentiment analysis.

Coming Soon

Computer Vision

Image classification, object detection, and segmentation with pre-trained models like ResNet, YOLO, and SAM.

In Planning

AutoML Engine

Automated hyperparameter tuning with Optuna, Ray Tune, and neural architecture search integration.

In Planning

Cloud Deployment

One-click deployment to AWS, GCP, and Azure with Docker and Kubernetes orchestration.

Long-term

LLM Integration

Large Language Model fine-tuning and inference with RAG support for custom knowledge bases.

Long-term

Multi-User Support

Team collaboration with user management, shared workspaces, and experiment versioning.