tf.errors.ResourceExhaustedError

Raised when some resource has been exhausted while running operation.

Inherits From: OpError

tf.errors.ResourceExhaustedError(
    node_def, op, message, *args
)

For example, this error might be raised if a per-user quota is exhausted, or perhaps the entire file system is out of space. If running into ResourceExhaustedError due to out of memory (OOM), try to use smaller batch size or reduce dimension size of model weights.

error_code The integer error code that describes the error.
experimental_payloads A dictionary describing the details of the error.
message The error message that describes the error.
node_def The NodeDef proto representing the op that failed.
op The operation that failed, if known.