Fruits-classification-120-different-fruits-

KO | EN

gitlite — search

#javascript #python #hacktoberfest #react #ai #typescript #llm #go #golang #android #machine-learning #rust #deep-learning #linux

Fruits-classification-120-different-fruits-

★ 9 Open GitHub ↗

Classified 120 different kinds of fruits using Deep Learning [Convolutional Neural Network(CNN), Transfer Learning, Deep Learning, Machine Learning]

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Fruits-classification-120-different-fruits-
Classified 120 different kinds of fruits using Deep Learning and machine learning approaches

Fruit Classification - Deep Learning Approach
Supermarkets around the world need to cluster different fruits to put them into their right racks and tag them with their right price. It is not easy especially nowadays when there are a lot of varieties of each fruit which differ very slightly with each other that even human eye can be confused.

To solve this problem, machine learning and deep learning techniques can be very useful as the development in these techniques especially in the field of neural networks has been immense in the last decade.

In this notebook, I will explore deep learning approaches which I will explore machine learning approaches in the other notebook. The dataset and problem has been taken from Kaggle. Here's the link for the same :

https://www.kaggle.com/moltean/fruits/tasks

Approaches Explored :-

1) Building convolutional layers along with maxpooling

2) Partial model with transfer learning approach from VGG16 and then adding some extra convolutional layers along with maxpooling

In the end, I will write what all things can be done later

I used google colab since it gives a free GPU and allow us to use the drive repository.

Dataset Properties

Total number of images: 82213.

Training set size: 61488 images (one fruit or vegetable per image).

Test set size: 20622 images (one fruit or vegetable per image).

Multi-fruits set size: 103 images (more than one fruit (or fruit class) per image)

Number of classes: 120 (fruits and vegetables).

Image size: 100x100 pixels.

First, I will mount my drive and use Kaggle's API to download data directly into the drive. It just takes a few secs to download around 800 MB of the fruits dataset

Here's my blog on this:

https://medium.com/@saketgarodia/image-classification-using-cnn-and-transfer-learning-approaches-8be8ffe4ff39?source=friends_link&sk=bc63af427ab151b8cda06ec3ad47fbe8

For Machine Learning approach, go through this link. Because of colab's computational issues, I used Kaggle kernel for this.

https://www.kaggle.com/saketgarodia1/kernel3d32a7bd5e?scriptVersionId=26482270

Fruits-classification-120-different-fruits-

Explore Similar Repositories

Was this content helpful?

Recent Feedback

Download README