search

PanJinquan / tensorflow_models_learning

tensorflow GoogleNet inception V1 V2 V3 V4
940 stars 289 forks source link

Performance issue in /slim/datasets (by P3) #30

Open DLPerf opened 3 years ago

DLPerf commented 3 years ago

Hello! I've found a performance issue in download_and_convert_cifar10.py: with tf.Session('') as sess(here) is defined in the function _add_to_tfrecord(here) which is repeatedly called in the loop for i in range(_NUM_TRAIN_FILES)(here).

tf.Session being defined repeatedly could lead to incremental overhead. If you define tf.Session out of the loop and pass tf.Session as a parameter to the loop, your program would be much more efficient. Here is the Stack Overflow post to support it.

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

DLPerf commented 3 years ago

Hello, I'm looking forward to your reply~