Model Curriculum
PYTHON WITH DEEP LEARNING
SECTOR: IT & ITeS
SUB-SECTOR: Data Science & AI
OCCUPATION: Deep Learning Engineer
REF ID: IT/ML/Q001
NSQF LEVEL: 5
1
Python with Deep Learning
CURRICULUM / SYLLABUS
This program is aimed at training candidates for the job of a "Deep Learning Engineer" in
the "IT & ITeS" sector/industry and aims to build the following key competencies:
Program Name Python with Deep Learning
Qualification Pack Name & IT/DL/Q001, v1.0
Reference ID
Version No. 1.0 Version Update Date NA
Pre-requisites to Training Basic knowledge of Python and machine learning.
Preferably a graduate or diploma in Computer Science/IT.
Training Outcomes After completing this program, participants will be able to:
Understand Deep Learning fundamentals: Learn and
apply neural network concepts.
Build and evaluate deep learning models: Implement
and train models using TensorFlow and Keras.
Explore Convolutional Neural Networks (CNNs) and
Recurrent Neural Networks (RNNs): Apply these to
computer vision and sequence tasks.
Deploy and scale models: Implement deep learning
solutions in production environments.
Course Structure
This course encompasses 4 National Occupational Standards (NOS) of "Deep Learning
Engineer" Qualification Pack issued by the IT Skill Council.
Sr. Module Key Learning Outcomes Equipment
No. Required
1 Introduction to Python basics: Variables, loops, Refer to Unique
Python and Deep functions equipment
Learning Data manipulation: NumPy, Required section
Pandas
Theory Duration Data visualization: Matplotlib,
(hh:mm) Seaborn
12:00
Practical Duration
(hh:mm)
15:00
2 Introduction to Overview of supervised and Refer to Unique
2
Machine Learning unsupervised learning equipment
Libraries: Scikit-learn for machine Required section
Theory Duration learning models
(hh:mm) Basic models: Linear regression,
15:00 classification (Logistic Regression,
Decision Trees)
Practical Duration
(hh:mm)
15:00
3 Introduction to Deep What is Deep Learning? Refer to Unique
Learning Overview of neural networks: equipment
Perceptron, multi-layer Required section
Theory Duration perceptron (MLP)
(hh:mm) Libraries: TensorFlow, Keras
15:00 basics
Activation functions, loss
Practical Duration functions, optimizers
(hh:mm)
15:00
4 Building Neural Constructing simple neural Refer to Unique
Networks networks in Keras equipment
Feedforward networks, Required section
Theory Duration backpropagation
(hh:mm) Training and evaluating models
12:00 Hyperparameters: learning rate,
epochs, batch size
Practical Duration
(hh:mm)
15:00
5 Convolutional Introduction to CNNs and their Refer to Unique
Neural Networks applications equipment
(CNN) Convolution layers, pooling Required section
layers, fully connected layers
Theory Duration Building CNNs for image
(hh:mm) classification (e.g., MNIST, CIFAR-
12:00 10)
Practical Duration
(hh:mm)
15:00
3
6 Recurrent Neural Introduction to sequence Refer to Unique
Network (RNN) modeling equipment
Understanding RNNs and their Required section
Theory Duration variations (LSTM, GRU)
(hh:mm) Building RNNs for time series data
12:00 or text classification
Practical Duration
(hh:mm)
15:00
7 Transfer Learning Concept of transfer learning in Refer to Unique
deep learning equipment
Theory Duration Pre-trained models: VGG, ResNet, Required section
(hh:mm) Inception
12:00 Fine-tuning and using pre-trained
models for custom tasks
Practical Duration
(hh:mm)
20:00
8 Advanced Topics in Autoencoders, Generative Refer to Unique
Deep Learning Adversarial Networks (GANs) equipment
Natural Language Processing Required section
Theory Duration (NLP) with deep learning: word
(hh:mm) embeddings, BERT
13:00 Reinforcement learning overview
Practical Duration
(hh:mm)
20:00
9 Model Deployment Saving and loading models
Model deployment with
Theory Duration TensorFlow Serving, Flask, or
(hh:mm) FastAPI
15:00 Introduction to cloud-based
deployment (AWS, Google Cloud)
Practical Duration
(hh:mm)
20:00
10 Capstone Project End-to-end deep learning project Refer to Unique
Data collection, preprocessing, equipment
4
Practical Duration building models, evaluating Required section
(hh:mm) performance
32:00 Presenting and deploying the
solution
Total Duration: Unique Equipment Required:
Theory Duration Laptop with at least 8 GB RAM
(hh:mm) Projector
130:00 Python IDE (Jupyter Notebook or PyCharm)
TensorFlow, Keras libraries installed
Practical Duration Access to datasets for machine learning
(hh:mm) Internet connection for deployment/cloud projects
170:00
Grand Total Course Duration: 300 Hours
Technical Outcome of Course:
Introduction to machine learning and deep learning
Libraries: TensorFlow, Keras | Laptop, Projector, Python IDE | 20:00 | 20:00 |
IT/DL/N0801 | | 2 | Building Neural Networks | - Understanding neural networks
and backpropagation
Building simple feedforward neural networks in Keras
Model training and evaluation | Laptop, Projector, Datasets | 25:00 | 35:00 |
IT/DL/N0802 | | 3 | Convolutional Neural Networks (CNNs) | - Introduction to CNNs
for image classification
Building CNNs using TensorFlow and Keras
Working with datasets like MNIST/CIFAR-10 | Laptop, Projector, Python IDE | 30:00 |
40:00 | IT/DL/N0803 | | 4 | Recurrent Neural Networks (RNNs) | - Introduction to
sequence models and RNNs
Implementing LSTM/GRU networks for text and time series data
Text generation and sequence prediction | Laptop, Projector, Python IDE | 30:00 |
40:00 | IT/DL/N0804 | | 5 | Advanced Deep Learning Topics | - Transfer learning
with pre-trained models (e.g., ResNet, BERT)
GANs and autoencoders for data generation
Deploying deep learning models in production | Laptop, Projector, Cloud Access |
25:00 | 35:00 | IT/DL/N0805 |
5
6
