You will get Deep Convolutional GAN (DCGAN) in Keras to generate pictures of bricks
Project details
If you are a teacher of ML or Gen AI, this project will help your students understand the DCGAN.
This project implemented DCGAN using bricks dataset
The project was well documented, and each step was commented on well, so you can easily explain the DCGAN to your students.
Steps of build DCGAN in Keras to generate pictures of bricks:
1. Build The Discriminator with 5 convolution layers with BatchNorm2d,
and LeakyReLU layers in between
2.Build The generator with 5 convolution transpose layers with BatchNorm2d,
and LeakyReLU layers in between
3. Build DCGAN using subclass of the abstract Keras Model class:
3.a. Build custom compile function to compute optimizer ,loss and
accuracy of discriminator and generator
3.b. Build custom metrics to monitor the custom training loop
3.c. Build custom training loop with TensorFlow’s Gradient Tape to compute gradients of operations executed during a
forward pass of a model
4. Create a custom callback that periodically saves generated images
5. By using ModelCheckpoint save the Keras model weights at some frequency
6. use Tensorboard in custom training loop to track the model performance
This project implemented DCGAN using bricks dataset
The project was well documented, and each step was commented on well, so you can easily explain the DCGAN to your students.
Steps of build DCGAN in Keras to generate pictures of bricks:
1. Build The Discriminator with 5 convolution layers with BatchNorm2d,
and LeakyReLU layers in between
2.Build The generator with 5 convolution transpose layers with BatchNorm2d,
and LeakyReLU layers in between
3. Build DCGAN using subclass of the abstract Keras Model class:
3.a. Build custom compile function to compute optimizer ,loss and
accuracy of discriminator and generator
3.b. Build custom metrics to monitor the custom training loop
3.c. Build custom training loop with TensorFlow’s Gradient Tape to compute gradients of operations executed during a
forward pass of a model
4. Create a custom callback that periodically saves generated images
5. By using ModelCheckpoint save the Keras model weights at some frequency
6. use Tensorboard in custom training loop to track the model performance
Machine Learning Tools
Keras, Python, TensorFlowWhat's included $150
These options are included with the project scope.
$150
- Delivery Time 3 days
- Number of Revisions 1
- Number of Model Variations 1
- Model Documentation
- Source Code
About Asmaa
Machine Learning Engineer | Data Analysis
Giza, Egypt - 11:44 pm local time
Whether you’re trying to make your desired work,list your services ,I will help!
I’m experienced in python,C++ so I could make many tasks , you wish
Steps for completing your project
After purchasing the project, send requirements so Asmaa can start the project.
Delivery time starts when Asmaa receives requirements from you.
Asmaa works on your project following the steps below.
Revisions may occur after the delivery date.
approve the dataset ,
approve the dataset , if you send it