Topic 1 Question 126
You are training an object detection model using a Cloud TPU v2. Training time is taking longer than expected. Based on this simplified trace obtained with a Cloud TPU profile, what action should you take to decrease training time in a cost-efficient way?
Move from Cloud TPU v2 to Cloud TPU v3 and increase batch size.
Move from Cloud TPU v2 to 8 NVIDIA V100 GPUs and increase batch size.
Rewrite your input function to resize and reshape the input images.
Rewrite your input function using parallel reads, parallel processing, and prefetch.
ユーザの投票
コメント(5)
- 正解だと思う選択肢: D
parallel reads, parallel processing, and prefetch is needed here
👍 4pshemol2022/12/20 - 正解だと思う選択肢: D👍 3hiromi2022/12/21
- 正解だと思 う選択肢: D
i didn't see v3 has any benefit than v2 https://cloud.google.com/tpu/docs/system-architecture-tpu-vm#performance_benefits_of_tpu_v3_over_v2
👍 1mymy94182022/12/17
シャッフルモード