Created by: cryoco

System information Paddle version: 1.8.2 Paddle With CUDA: False OS: Ubuntu 16.04 CPU: 16 Intel(R) Xeon(R) CPU E5-2620 v3 @ 2.40GHz Python version: 3.5.2 CUDA version: 9.0.176 cuDNN version: None.None.None Nvidia driver version: None API information: inference configuration

config.disable_gpu()
config.enable_mkldnn()
config.set_cpu_math_library_num_threads(4)

To Reproduce

1. download models and data

• ocr detection model
• ocr recognition model
• test data The test data pictures are binary files saved with np.save, and can be loaded with np.load.

2. perform inference with mkldnn code to reproduce: test_ocr_mkldnn_mem.txt(rename to test_ocr_mkldnn_mem.py)

• detection

python3 test_ocr_mkldnn_mem.py --model_file=./ch_det_mv3_db/model --params_file=./ch_det_mv3_db/params --mode=det --mkldnn

• recognition

python3 test_ocr_mkldnn_mem.py --model_file=./ch_rec_mv3_crnn/model --params_file=./ch_rec_mv3_crnn/params --mode=rec --mkldnn

Memory usage increasing can be witnessed with top command.

3. perform inference without mkldnn

• detection

python3 test_ocr_mkldnn_mem.py --model_file=./ch_det_mv3_db/model --params_file=./ch_det_mv3_db/params --mode=det 

• recognition

python3 test_ocr_mkldnn_mem.py --model_file=./ch_rec_mv3_crnn/model --params_file=./ch_rec_mv3_crnn/params --mode=rec

Memory usage will maintain stable.

The input shapes we use in OCR are dynamic, which might be relevant to this issue.