COMPARING OPTIMIZER ADAM, RMSPROP, AND SGD IN THE CLASSIFICATION OF BANANA RIPENESS USING CNN ALGORITHM

GUNAWAN, IRVAN (2021) COMPARING OPTIMIZER ADAM, RMSPROP, AND SGD IN THE CLASSIFICATION OF BANANA RIPENESS USING CNN ALGORITHM. Other thesis, Universitas Katholik Soegijapranata Semarang.

[img] Text
17.K1.0024-IRVAN GUNAWAN-COVER_a.pdf
Download (1MB)
[img] Text
17.K1.0024-IRVAN GUNAWAN-BAB I_a.pdf
Download (227kB)
[img] Text
17.K1.0024-IRVAN GUNAWAN-BAB II_a.pdf
Restricted to Registered users only
Download (229kB)
[img] Text
17.K1.0024-IRVAN GUNAWAN-BAB III_a.pdf
Download (225kB)
[img] Text
17.K1.0024-IRVAN GUNAWAN-BAB IV_a.pdf
Download (147kB)
[img] Text
17.K1.0024-IRVAN GUNAWAN-BAB V_a.pdf
Download (325kB)
[img] Text
17.K1.0024-IRVAN GUNAWAN-BAB VI_a.pdf
Download (109kB)
[img] Text
17.K1.0024-IRVAN GUNAWAN-DAPUS_a.pdf
Download (328kB)
[img] Text
17.K1.0024-IRVAN GUNAWAN-LAMP_a.pdf
Download (225kB)

Abstract

Banana is one of the most popular fruit in indonesia because this fruit is very easy to find in every area and the price offered is also relatively very affordable. But for some people don’t know the level of ripeness of banana . In this project will classification the level ripeness of banana. Ripeness banana have different colors, so we need a method that can classification the level of maturiry . Therefore, we use Convolutional Neural Network method which can be called CNN, and will compare 3 optimizer, namely Adam, RMSprop, and SGD, and compare 3 result with different optimizer to find out which optimizer is better for this project. The final result for this project is that CNN can run well for classification of banana ripeness and the predicting is also quite good, this project get a final accuracy result 93,75% using SGD optimizer for classification the ripeness level of banana fruit.

Item Type: Thesis (Other)
Subjects: 000 Computer Science, Information and General Works > 004 Data processing & computer science
Divisions: Faculty of Computer Science > Department of Informatics Engineering
Depositing User: mr AM. Pudja Adjie Sudoso
Date Deposited: 15 Oct 2021 02:49
Last Modified: 15 Oct 2021 02:49
URI: http://repository.unika.ac.id/id/eprint/27129

Actions (login required)

View Item View Item
Unika Repository is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.