Skip to content

Bangkit-Capstone-Project-Team/MachineLearning

Repository files navigation

BATIK CLASSIFICATION WITH MOBILNETS ARCHITECTURE

In this project, I implemented a deep learning architecture Mobilenet to identification and classification batik to 20 classes (batik-bali, batik-betawi, batik-celup, batik-cendrawasih, batik-ceplok, batik-ciamis, batik-garutan, batik-gentongan, batik-kawung, batik-keraton, batik-lasem, batik-megamendung, batik-parang, batik-pekalongan, batik-priangan, batik-sekar, batik-sidoluhur, batik-sidomukti, batik-sogan, batik-tambal)

For doing this project we used the following resources: https://www.kaggle.com/dionisiusdh/indonesian-batik-motifs

Model specifications and results

The architecture of this project is based on Mobilnet.

we use a pre-trained model Because the pre-trained models have their parameters more adjusted and the filters they have learned are more polished, in consequence, we can avoid the process of teaching the network those filters and the only thing that we should focus on is to adjust the other parameters to the specifications of our data set.

#How to use

you need to instal :

!pip install tflite-model-maker (for get all requirment) and you'll get:

  • keras-preprocessing=1.1.2
  • tensorboard=2.5
  • numpy==1.16.1
  • opencv-python==4.0.0.21
  • scikit-learn=0.19.0
  • python=3.7

Download dataset and split it to be 3 folder

  • data_training
  • data_testing
  • validation

Augumented Data

Training prosess

train

Evaluate Model

acc

Results

  • PREDICT LABEL result

the last is save your model

save model

About

klasifikasi batik berdasarkan nama daerah

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published