You are on page 1of 21

Big Data

Practice
Diesty Zakyah Prayogi (1401201595)
Dimas Rizqie Zanetti (1401203104)
Rizky Fadilyulian (1401203583)
CONVOLUTIONAL NEURAL NETWORK
Import Library

Import the required libraries


Import Dataset CIFAR-10

Importing the Cifar-10 Dataset to be used in the model we will be working


with
Pemodelan

Modeling the classes that will be used in the CIFAR-10 dataset


Data Exploration

Then we carry out data exploration on the imported CIFAR-10 dataset


Build Model
Define Model

Next we build the model with the first step using define model
Build Model
Fitting Model

After defining the model, we do the


model fitting to determine how many
times the learning algorithm will work to
process the training and epoch datasets
Build Model
Testing Model

Then we do model testing to get an accuracy test value whose result is 64%.
Predict Image

Then do predict images to get plotting first x test


images , prediction labels , and actual labels
Trainng Model

Doing model training using horse images. This aims to test the
capabilities of the model
Tugas

To complete the task, we try to use a car image as


a test object
Tugas

The following is the result of model


processing of the car images we use.
Here the model can correctly classify
the image as automobile/vehicle/car.
SIMPLE DEEPFAKE LIPSYNC
Install Wav2Lip

Install and import Wav2Lip


Mengimport video and audio dari data set yang
diberikan

Import video and audio from the dataset to be


used
Mengimport video and audio dari data set yang
diberikan

Import video and audio from the dataset to be used


Processing

Here the video is processed with audio using the deepfake


model, namely the wav2lip model
Scalling 50%

After that the video is scaled 50% to


minimize file size
Downloading file

Download the pre-processed file


Thankyou

You might also like