This is image data of Natural Scenes around the world. Containing 6 different class
This Data contains around 25k images of size 150x150 distributed under 6 categories. {'buildings' -> 0, 'forest' -> 1, 'glacier' -> 2, 'mountain' -> 3, 'sea' -> 4, 'street' -> 5 }
I got this data from Kaggle Dataset : https://www.kaggle.com/puneet6060/intel-image-classification
Notes : I make this program using google colaboratory, so the way i handled files is based on colab's way. First i've to upload the file to my Drive then extract the file.
The train Accuracy i got is 90.6% after 35 Epoch which i believe could be higher if i tweak the model.
I you've any suggestion according to this deeplearning program please tell me :)