the dataset you provided wasn't sorted in the directories like your dataset is. Have I missed something?

thanks a lot for sharing

Ma’am I have a doubt is this run on web

I have used same dataset.but after pre processing I am getting different no. of images.

Why no softmax layer?, Why have you split the train data into train and validation when test was seperately given

In your program image_count_train = len(list(data_dir_train.glob('*/*.jpg'))) print(image_count_train) image_count_test = len(list(data_dir_test.glob('*/*.jpg'))) print(image_count_test) 2239 118 train_ds = tf.keras.preprocessing.image_dataset_from_directory( data_dir_train, seed=123, validation_split = 0.2, subset='training', image_size=(img_height, img_width), batch_size=batch_size) Found 2239 files belonging to 9 classes. Using 1792 files for training. but when I run the same code, I get the same train and test data count.After preprocessing output is different from yours code output why it is so plz tell me . image_count_train = len(list(data_dir_train.glob('*/*.jpg'))) print(image_count_train) image_count_test = len(list(data_dir_test.glob('*/*.jpg'))) print(image_count_test) 2239 120 train_ds = tf.keras.preprocessing.image_dataset_from_directory( data_dir_train, seed=123, validation_split = 0.3, subset='training', image_size=(180, 180), batch_size=32) Found 29183 files belonging to 9 classes. Using 20429 files for training. anyone other who is implementing this code is facing same problem.

data set is very big any other dataset can we use?

I am not able to access your dataset, can you tell me how to get this

For me also I am getting less validation accuracy Thanks for making this video madam

is there the same project using pca

which parameters you can take