Data formatting for 1D-CNN in R - r

Question

I have an array of data (around 1000+ cases (rows), and about 650+ columns of information (V1 - V5 in the example below) which I want to use for a keras 1D-CNN (which has 3 classifications (class in the example below)).
I'm having trouble trying to format the data in a way that will fit the model that I have written.
The data looks similar to this:
set.seed(999)
case <- c('A','B','C','D','E','F','G','H')
class <- c('N','N','T','R','T','R','N','T')
V1 <- runif(8, min = 0, max = 1)
V2 <- runif(8, min = 0, max = 1)
V3 <- runif(8, min = 0, max = 1)
V4 <- runif(8, min = 0, max = 1)
V5 <- runif(8, min = 0, max = 1)
df <- data.frame(case, class, V1, V2, V3, V4, V5)
df
The keras 1D-CNN I have written is this:
build.model <- function() {
model <- keras_model_sequential()
model %>%
layer_conv_1d(
filter = 32, kernel_size = c(5), padding = "same",
input_shape = c(100, 3)
) %>%
layer_activation("relu") %>%
# Use max pooling
layer_max_pooling_1d(pool_size = c(2)) %>%
# Flatten max filtered output into feature vector
# and feed into dense layer
layer_flatten() %>%
layer_dense(128) %>%
layer_activation('relu') %>%
layer_dense(3) %>%
layer_activation('softmax')
model %>% compile(
loss = 'categorical_crossentropy',
optimizer = 'rmsprop',
metrics = 'accuracy'
)
return(model)
}
model <- build.model()
print(model)
When I try to pass the df through the model it doesn't work.
I know it's to do with the format of the data, but I'm at a loss as to how to rectify this.
Any tips/pointers would be greatly appreciated.
TIA
Miki