about dataset_map

[zouw2]

from the intro: "Note that while dataset_map() is defined using an R function, there are some special constraints on this function which allow it to execute not within R but rather within the TensorFlow graph."

Does the 'constraint' mean we can only use certain functions available to tf within dataset_map? If we call certain R library functions in side of dataset_map, will this work?

I am looking for data augment (using R functions) examples that work in eager execute style. thanks wei

[t-kalinowski]

Another way to phrase it is: only operations on tensors are recorded (aka traced) by the function when the dataset pipeline is built. Then when the pipeline is run, only the recorded tensor operations are rerun. Meaning, that only the tensor operations are meaningful in the dataset_map() context.