Ndtyjky

I am new to image classification.
I am using CNN to classify 3 images of 3 products.
Below is my code -

model <- keras_model_sequential()



model %>%



     layer_conv_2d(filters = 64, 

                   kernel_size = c(3,3),

                   activation = 'relu',

                   input_shape = c(100,100,3)) %>% 



   layer_conv_2d(filters = 32,

                   kernel_size = c(3,3),

                   activation = 'relu') %>%

layer_max_pooling_2d(pool_size = c(2,2)) %>%



     layer_dropout(rate = 0.25) %>%



     layer_conv_2d(filters = 64,

                   kernel_size = c(3,3),

                   activation = 'relu') %>%



     layer_conv_2d(filters = 64,

                   kernel_size = c(3,3),

                   activation = 'relu') %>%

     layer_max_pooling_2d(pool_size = c(2,2)) %>%

     layer_dropout(rate = 0.25) %>%

      layer_flatten() %>%

     layer_dense(units = 256, activation = 'relu') %>%

     layer_dropout(rate=0.25) %>%

     layer_dense(units = 3, activation = 'softmax') %>%



     compile(loss = 'categorical_crossentropy',

             optimizer = optimizer_sgd(lr = 0.01,

                                       decay = 1e-6,

                                       momentum = 0.9,

                                       nesterov = T),

             metrics = c('accuracy'))

# Fit model

history <- model %>%

     fit(train,

         trainLabels,

         epochs = 60,

         batch_size =  32,

         validation_split = 0.2,

         validation_data = list(test, testLabels))

The accuracy that i am getting from this is 0.333
i would like to improve this accuracy. i am not able to find any tutorials which explains on how to decide the appropriate kernel size or the appropriate no of filters.
Can someone explain same to me?
Please let me know if you require any other details.

Thanks in advance