Performance issues in data.py(P2)

[DLPerf]

Hello,I found a performance issue in the definition of load_dataset , data.py, dataset = dataset.map(_parser) was called without num_parallel_calls. I think it will increase the efficiency of your program if you add this.

The same issues also exist in dataset = dataset.map(_parser) , dataset = dataset.map(_preprocess_inference), dataset = dataset.map(_parser), dataset = dataset.map(_parser) and dataset = dataset.map(_preprocess_inference)

Here is the documemtation of tensorflow to support this thing.

Looking forward to your reply. Btw, I am very glad to create a PR to fix it if you are too busy.