[DM/DL] 텐서플로우의 인셉션v3 실행하기 (4)

우선 실행하게 되면 CPU가 하늘 높은 줄 모르고 열심히 일하는 것을 볼수 있을 것이다.

우잉잉..소리도 나고...예전에 CPU 성능이 부족할때 기분을 한껏 느낄수 있다. 

실행시에 성능에 따라 learning_rate와 batch_size를 조절할수 있고 max_steps을 조정하여 전체 시간을 줄일수 있다. gpu는...일반 환경이니 굳이 늘릴 필요가 없을 것으로 보여지고...

root@hoonjo-VirtualBox:~/inception/models/inception# time bazel-bin/inception/flowers_train --train_dir=$FLOWERS_DIR/train --data_dir=$FLOWERS_DIR/data --pretrained_model_checkpoi 
nt_path=$INCEPTION_DIR/inception-v3/model.ckpt-157585 --fine_tune=True --initial_learning_rate=0.001 -input_queue_memory_factor=1 --max_steps=50 --num_gpus 1 --batch_size=32
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE3 instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.1 instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.2 instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX instructions, but these are available on your machine and could speed up CPU computations.
2017-06-02 08:56:53.061730: Pre-trained model restored from /root/inception/inception-v3/model.ckpt-157585
2017-06-02 08:59:30.940674: step 0, loss = 2.78 (0.4 examples/sec; 74.705 sec/batch)
2017-06-02 09:07:18.404594: step 10, loss = 2.60 (0.9 examples/sec; 35.722 sec/batch)
2017-06-02 09:13:06.446650: step 20, loss = 2.41 (0.9 examples/sec; 34.499 sec/batch)
2017-06-02 09:18:51.787365: step 30, loss = 2.17 (0.9 examples/sec; 34.545 sec/batch)
2017-06-02 09:24:44.551172: step 40, loss = 2.07 (0.9 examples/sec; 34.955 sec/batch)

real 35m29.532s
user 213m28.596s
sys 19m26.928s

한개만 하면 재미가 없으니 다른 것을 실습해 보도록 하자. 
위에 것이 끝나기를 30분이 넘게 기다리고 있다....그래야 다음 것을 할 것이니까..

이 내용만 봐서는 CPU는 많이 쓰는건 알겠는데 뭐하는지 알기가 어렵다..
따라서, 좀더 쓸만한 예제가 필요하다..!!!

To be continue....