# rupakraj/deeplearning architecture

> Architecture diagram of the GitHub repository rupakraj/deeplearning, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-11-16). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/rupakraj/deeplearning
- Repository: https://github.com/rupakraj/deeplearning

## Overview

DeepLearning is a classroom collection of runnable Jupyter lessons, not a single application pipeline. The principal workflow is for a learner to open a topic notebook, select a Python/Jupyter kernel, and run cells to explore its demonstrations. The graph groups the documented lessons by subject and includes the separate representation-learning collection and its synthetic CRM-data generator. Lesson capabilities come from the README; notebook internals were not sampled, so no connections between lessons or to datasets are inferred.

Main entry points: [`rep-learning/scripts/make_crm.py`](https://github.com/rupakraj/deeplearning/blob/main/rep-learning/scripts/make_crm.py) (Synthetic CRM Generator), [`pytorch_toy_example.ipynb`](https://github.com/rupakraj/deeplearning/blob/main/pytorch_toy_example.ipynb) (PyTorch Toy Example), [`overfit_underfit_regularization.ipynb`](https://github.com/rupakraj/deeplearning/blob/main/overfit_underfit_regularization.ipynb) (Regularization), [`limitation_of_gradient.ipynb`](https://github.com/rupakraj/deeplearning/blob/main/limitation_of_gradient.ipynb) (Gradient Descent Limits)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_foundations["Foundations &amp; Optimization"]
  node_pytorch_intro["PyTorch Toy Example"]
  node_regularization["Regularization"]
  node_gradient_limits["Gradient Descent Limits"]
  node_learning_rates["Adaptive Learning Rates"]
end

subgraph group_vision["Computer Vision"]
  node_cnn_motivation["CNN Motivation<br/>[need_for_cnn.ipynb]"]
  node_alexnet["AlexNet Classification"]
  node_batch_norm["Batch Normalization"]
end

subgraph group_representation["Representation Learning"]
  node_autoencoder["Autoencoder Representations"]
  node_word2vec["Word2Vec Representations"]
  node_rep_dataset["Representation Dataset Prep"]
  node_rep_training["Representation Model Training"]
end

subgraph group_language["Language &amp; Retrieval"]
  node_embedding_retrieval["Embedding Retrieval"]
  node_sequence_models["Sequence Models"]
  node_ngram_model["Neural N-Gram Model"]
  node_embedding_building["Embedding Building"]
end

subgraph group_crm["CRM Data"]
  node_crm_generator["Synthetic CRM Generator<br/>[make_crm.py]"]
  node_crm_data[("CRM Dataset<br/>[crm.csv]")]
end

node_learner(("Learner"))
node_notebook_runtime["Jupyter / Python Runtime"]

node_learner -->|"opens and runs"| node_pytorch_intro
node_learner -->|"opens and runs"| node_regularization
node_learner -->|"opens and runs"| node_gradient_limits
node_learner -->|"opens and runs"| node_learning_rates
node_learner -->|"opens and runs"| node_cnn_motivation
node_learner -->|"opens and runs"| node_alexnet
node_learner -->|"opens and runs"| node_batch_norm
node_learner -->|"opens and runs"| node_autoencoder
node_learner -->|"opens and runs"| node_word2vec
node_learner -->|"opens and runs"| node_embedding_retrieval
node_learner -->|"opens and runs"| node_sequence_models
node_learner -->|"opens and runs"| node_ngram_model
node_crm_generator -->|"writes CSV"| node_crm_data

click node_pytorch_intro "https://github.com/rupakraj/deeplearning/blob/main/pytorch_toy_example.ipynb"
click node_regularization "https://github.com/rupakraj/deeplearning/blob/main/overfit_underfit_regularization.ipynb"
click node_gradient_limits "https://github.com/rupakraj/deeplearning/blob/main/limitation_of_gradient.ipynb"
click node_learning_rates "https://github.com/rupakraj/deeplearning/blob/main/adaptive_learning_rates.ipynb"
click node_cnn_motivation "https://github.com/rupakraj/deeplearning/blob/main/need_for_cnn.ipynb"
click node_alexnet "https://github.com/rupakraj/deeplearning/blob/main/alexnet_image_classification.ipynb"
click node_batch_norm "https://github.com/rupakraj/deeplearning/blob/main/batch_normalization.ipynb"
click node_autoencoder "https://github.com/rupakraj/deeplearning/blob/main/autoencoder_representation_learning.ipynb"
click node_word2vec "https://github.com/rupakraj/deeplearning/blob/main/word2vec_autoencoder.ipynb"
click node_rep_dataset "https://github.com/rupakraj/deeplearning/blob/main/rep-learning/01_dataset_preparation.ipynb"
click node_rep_training "https://github.com/rupakraj/deeplearning/blob/main/rep-learning/02_model_and_training.ipynb"
click node_embedding_retrieval "https://github.com/rupakraj/deeplearning/blob/main/embedding_retrieval.ipynb"
click node_sequence_models "https://github.com/rupakraj/deeplearning/blob/main/need_for_sequence_models.ipynb"
click node_ngram_model "https://github.com/rupakraj/deeplearning/blob/main/neural_ngram_language_model.ipynb"
click node_embedding_building "https://github.com/rupakraj/deeplearning/blob/main/rep-learning/building_embedding.ipynb"
click node_crm_generator "https://github.com/rupakraj/deeplearning/blob/main/rep-learning/scripts/make_crm.py"
click node_crm_data "https://github.com/rupakraj/deeplearning/blob/main/rep-learning/data/crm.csv"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_pytorch_intro,node_regularization,node_gradient_limits,node_learning_rates toneBlue
class node_cnn_motivation,node_alexnet,node_batch_norm toneAmber
class node_autoencoder,node_word2vec,node_rep_dataset,node_rep_training toneMint
class node_embedding_retrieval,node_sequence_models,node_ngram_model,node_embedding_building toneRose
class node_crm_generator,node_crm_data,node_learner,node_notebook_runtime toneIndigo
```

## Components

### Foundations & Optimization

- [PyTorch Toy Example](https://github.com/rupakraj/deeplearning/blob/main/pytorch_toy_example.ipynb): `pytorch_toy_example.ipynb`
- [Regularization](https://github.com/rupakraj/deeplearning/blob/main/overfit_underfit_regularization.ipynb): `overfit_underfit_regularization.ipynb`
- [Gradient Descent Limits](https://github.com/rupakraj/deeplearning/blob/main/limitation_of_gradient.ipynb): `limitation_of_gradient.ipynb`
- [Adaptive Learning Rates](https://github.com/rupakraj/deeplearning/blob/main/adaptive_learning_rates.ipynb): `adaptive_learning_rates.ipynb`

### Computer Vision

- [CNN Motivation](https://github.com/rupakraj/deeplearning/blob/main/need_for_cnn.ipynb): `need_for_cnn.ipynb`
- [AlexNet Classification](https://github.com/rupakraj/deeplearning/blob/main/alexnet_image_classification.ipynb): `alexnet_image_classification.ipynb`
- [Batch Normalization](https://github.com/rupakraj/deeplearning/blob/main/batch_normalization.ipynb): `batch_normalization.ipynb`

### Representation Learning

- [Autoencoder Representations](https://github.com/rupakraj/deeplearning/blob/main/autoencoder_representation_learning.ipynb): `autoencoder_representation_learning.ipynb`
- [Word2Vec Representations](https://github.com/rupakraj/deeplearning/blob/main/word2vec_autoencoder.ipynb): `word2vec_autoencoder.ipynb`
- [Representation Dataset Prep](https://github.com/rupakraj/deeplearning/blob/main/rep-learning/01_dataset_preparation.ipynb): `rep-learning/01_dataset_preparation.ipynb`
- [Representation Model Training](https://github.com/rupakraj/deeplearning/blob/main/rep-learning/02_model_and_training.ipynb): `rep-learning/02_model_and_training.ipynb`

### Language & Retrieval

- [Embedding Retrieval](https://github.com/rupakraj/deeplearning/blob/main/embedding_retrieval.ipynb): `embedding_retrieval.ipynb`
- [Sequence Models](https://github.com/rupakraj/deeplearning/blob/main/need_for_sequence_models.ipynb): `need_for_sequence_models.ipynb`
- [Neural N-Gram Model](https://github.com/rupakraj/deeplearning/blob/main/neural_ngram_language_model.ipynb): `neural_ngram_language_model.ipynb`
- [Embedding Building](https://github.com/rupakraj/deeplearning/blob/main/rep-learning/building_embedding.ipynb): `rep-learning/building_embedding.ipynb`

### CRM Data

- [Synthetic CRM Generator](https://github.com/rupakraj/deeplearning/blob/main/rep-learning/scripts/make_crm.py): `rep-learning/scripts/make_crm.py`
- [CRM Dataset](https://github.com/rupakraj/deeplearning/blob/main/rep-learning/data/crm.csv): `rep-learning/data/crm.csv`

### Other components

- **Learner**
- **Jupyter / Python Runtime**

## Connections

- Learner → PyTorch Toy Example: opens and runs (evidence: `readme.md`)
- Learner → Regularization: opens and runs (evidence: `readme.md`)
- Learner → Gradient Descent Limits: opens and runs (evidence: `readme.md`)
- Learner → Adaptive Learning Rates: opens and runs (evidence: `readme.md`)
- Learner → CNN Motivation: opens and runs (evidence: `readme.md`)
- Learner → AlexNet Classification: opens and runs (evidence: `readme.md`)
- Learner → Batch Normalization: opens and runs (evidence: `readme.md`)
- Learner → Autoencoder Representations: opens and runs (evidence: `readme.md`)
- Learner → Word2Vec Representations: opens and runs (evidence: `readme.md`)
- Learner → Embedding Retrieval: opens and runs (evidence: `readme.md`)
- Learner → Sequence Models: opens and runs (evidence: `readme.md`)
- Learner → Neural N-Gram Model: opens and runs (evidence: `readme.md`)
- Synthetic CRM Generator → CRM Dataset: writes CSV (evidence: `rep-learning/scripts/make_crm.py`)
