Image recognition using tensorflow and teras in R - error

Machine Learning and Modeling
, ,
1

I am following this tutorial - How to build your own image recognition app with R! [Part 1] | R-bloggers

But have come across an error.

model_function <- function(learning_rate = 0.001, 
                           dropoutrate=0.2, n_dense=1024){
  
  k_clear_session()
  
  model <- keras_model_sequential() %>%
    mod_base %>% 
    layer_global_average_pooling_2d() %>% 
    layer_dense(units = n_dense) %>%
    layer_activation("relu") %>%
    layer_dropout(dropoutrate) %>%
    layer_dense(units=output_n, activation="softmax")
  
  model %>% compile(
    loss = "categorical_crossentropy",
    optimizer = optimizer_adam(lr = learning_rate),
    metrics = "accuracy"
  )
  
  return(model)
  
}

model <- model_function()

The last line of this code gets the following error -

Error in py_call_impl(callable, call_args$unnamed, call_args$named) : 
  ValueError: Inputs to a layer should be tensors. Got '<Sequential name=sequential, built=False>' (of type <class 'keras.src.models.sequential.Sequential'>) as input for layer 'xception'.
Run `reticulate::py_last_error()` for details.

Does anyone know how I can fix this? Or have any recommendations on a different tutorial to follow?