aerial-cactus-identification

It's a Kaggle compittion to detect the cactus from aerial images

Getting Started

it's a classifier which classify images to have cactus and no cactus

achievements

got a 99.5% valid accurcy and 0.018 valid loss with just 15 epochs

100% of the public data classified well

private data classification

4000 images with no labels classified well by 100%

use this code for testing

Explain what these tests test and why

df.head()
df2=np.asarray(df)

testing=df2[20,:]

img_path = "/content/test/"+testing[0]
img = image.load_img(img_path, target_size=(32, 32))
#print(type(img))
plt.imshow(img)
print(testing[1])

Authors

• Sayed mohamed - Initial work - sayedmohamed

License

This project is licensed under the MIT License - see the LICENSE.md file for details