I want to run the MMS docker container serving the ArcFace-ResNet100 model. So I ran the following command to use the archived model that exists in the model zoo.
docker run -it --name mms -p 8080:8080 -p 8081:8081 awsdeeplearningteam/multi-model-server multi-model-server --start --models r100=https://s3.amazonaws.com/model-server/model_archive_1.0/onnx-arcface-resnet100.mar
But I got an import error.
2020-11-02 10:42:18,589 [INFO ] W-9000-r100-stdout com.amazonaws.ml.mms.wlm.WorkerLifeCycle - Traceback (most recent call last):
2020-11-02 10:42:18,589 [INFO ] W-9000-r100-stdout com.amazonaws.ml.mms.wlm.WorkerLifeCycle - File "/usr/local/lib/python3.6/dist-packages/mms/model_service_worker.py", line 1
74, in start_worker
2020-11-02 10:42:18,589 [INFO ] W-9000-r100-stdout com.amazonaws.ml.mms.wlm.WorkerLifeCycle - self.handle_connection(cl_socket)
2020-11-02 10:42:18,589 [INFO ] W-9000-r100-stdout com.amazonaws.ml.mms.wlm.WorkerLifeCycle - File "/usr/local/lib/python3.6/dist-packages/mms/model_service_worker.py", line 1
43, in handle_connection
2020-11-02 10:42:18,589 [INFO ] W-9000-r100-stdout com.amazonaws.ml.mms.wlm.WorkerLifeCycle - result, code = self.load_model(msg)
2020-11-02 10:42:18,589 [INFO ] W-9000-r100-stdout com.amazonaws.ml.mms.wlm.WorkerLifeCycle - File "/usr/local/lib/python3.6/dist-packages/mms/model_service_worker.py", line 1
09, in load_model
2020-11-02 10:42:18,589 [INFO ] W-9000-r100-stdout com.amazonaws.ml.mms.wlm.WorkerLifeCycle - self.service = self.model_loader.load(model_name, model_dir, handler, gpu, batc
h_size)
2020-11-02 10:42:18,589 [INFO ] W-9000-r100-stdout com.amazonaws.ml.mms.wlm.WorkerLifeCycle - File "/usr/local/lib/python3.6/dist-packages/mms/model_loader.py", line 116, in l
oad
2020-11-02 10:42:18,589 [INFO ] W-9000-r100-stdout com.amazonaws.ml.mms.wlm.WorkerLifeCycle - self.module = importlib.import_module(module_name)
2020-11-02 10:42:18,589 [INFO ] W-9000-r100-stdout com.amazonaws.ml.mms.wlm.WorkerLifeCycle - File "/usr/lib/python3.6/importlib/__init__.py", line 126, in import_module
2020-11-02 10:42:18,589 [INFO ] W-9000-r100-stdout com.amazonaws.ml.mms.wlm.WorkerLifeCycle - return _bootstrap._gcd_import(name[level:], package, level)
2020-11-02 10:42:18,589 [INFO ] W-9000-r100-stdout com.amazonaws.ml.mms.wlm.WorkerLifeCycle - File "<frozen importlib._bootstrap>", line 994, in _gcd_import
2020-11-02 10:42:18,589 [INFO ] W-9000-r100-stdout com.amazonaws.ml.mms.wlm.WorkerLifeCycle - File "<frozen importlib._bootstrap>", line 971, in _find_and_load
2020-11-02 10:42:18,589 [INFO ] W-9000-r100-stdout com.amazonaws.ml.mms.wlm.WorkerLifeCycle - File "<frozen importlib._bootstrap>", line 955, in _find_and_load_unlocked
2020-11-02 10:42:18,589 [INFO ] W-9000-r100-stdout com.amazonaws.ml.mms.wlm.WorkerLifeCycle - File "<frozen importlib._bootstrap>", line 665, in _load_unlocked
2020-11-02 10:42:18,589 [INFO ] W-9000-r100-stdout com.amazonaws.ml.mms.wlm.WorkerLifeCycle - File "<frozen importlib._bootstrap_external>", line 678, in exec_module
2020-11-02 10:42:18,589 [INFO ] W-9000-r100-stdout com.amazonaws.ml.mms.wlm.WorkerLifeCycle - File "<frozen importlib._bootstrap>", line 219, in _call_with_frames_removed
2020-11-02 10:42:18,589 [INFO ] W-9000-r100-stdout com.amazonaws.ml.mms.wlm.WorkerLifeCycle - File "/home/model-server/tmp/models/550bcc6f600b75976dc9d4353909df09-15/arcface_s
ervice.py", line 21, in <module>
2020-11-02 10:42:18,589 [INFO ] W-9000-r100-stdout com.amazonaws.ml.mms.wlm.WorkerLifeCycle - import cv2
2020-11-02 10:42:18,589 [INFO ] W-9000-r100-stdout com.amazonaws.ml.mms.wlm.WorkerLifeCycle - File "/home/model-server/.local/lib/python3.6/site-packages/cv2/__init__.py", lin
e 5, in <module>
2020-11-02 10:42:18,589 [INFO ] W-9000-r100-stdout com.amazonaws.ml.mms.wlm.WorkerLifeCycle - from .cv2 import *
2020-11-02 10:42:18,589 [INFO ] W-9000-r100-stdout com.amazonaws.ml.mms.wlm.WorkerLifeCycle - ImportError: libGL.so.1: cannot open shared object file: No such file or directory
Is this problem related to the archived model? Should I create a new docker image from awsdeeplearningteam/multi-model-server and install the dependencies?
I want to run the MMS docker container serving the ArcFace-ResNet100 model. So I ran the following command to use the archived model that exists in the model zoo.
docker run -it --name mms -p 8080:8080 -p 8081:8081 awsdeeplearningteam/multi-model-server multi-model-server --start --models r100=https://s3.amazonaws.com/model-server/model_archive_1.0/onnx-arcface-resnet100.mar
But I got an import error.Is this problem related to the archived model? Should I create a new docker image from
awsdeeplearningteam/multi-model-server
and install the dependencies?