awslabs/multi-model-server

Issues

• Make healthcheck dependent on models loaded

  #1027 opened 8 months ago by wojciechrauk-plutoflume
  1

• Streaming support on MMS

  #1012 opened 2 years ago by rauldiaz
  1

• Issue: Memory Leak when serving multiple models

  #999 opened 3 years ago by pratikluitel
  4

• MMS Server getting stuck while registering the model and says "worker pid is not available yet"

  #1025 opened a year ago by suchith-sixsense
  0

• How to handle invalid input!

  #1022 opened a year ago by xuweidongkobe
  1

• Improve MMS model loading exception handling

  #1010 opened 2 years ago by namannandan
  0

• Overriding the model routing logic

  #1017 opened 2 years ago by James-UnlikelyAI
  0

• readAddress(..) failed: Connection reset by peer

  #1016 opened 2 years ago by xuweidongkobe
  1

• Update documentation to establish the difference between backend time and backend response time

  #1014 opened 2 years ago by sachanub
  0

• Config option to increase or disable model load timeout

  #1009 opened 2 years ago by svenkata9
  1

• Permission denied when loading model

  #949 opened 4 years ago by akulk314
  1

• Invalid Response Headers Set for Non MME Inference Scenario (scikit learn container)

  #1007 opened 2 years ago by Grassycup
  0

• python version

  #1003 opened 2 years ago by n0thing233
  1

• Validate MMS process during start

  #1005 opened 2 years ago by nikhil-sk
  0

• Model specific custom python package installation

  #1004 opened 2 years ago by salvadiswar02
  0

• Preloading models on Sagemaker multi-model endpoint doesn't work

  #1001 opened 2 years ago by sassarini-marco
  0

• How to achieve autoscaling when running MMS on a fargate?

  #970 opened 4 years ago by sunilkumarmohanty
  1

• Is it supporting Apple M1 Chip ?

  #1000 opened 3 years ago by jaiswalvineet
  0

• mms-gpu - cuda error - No kernel image is available for execution on the device

  #998 opened 3 years ago by kaushal-idx
  0

• how to change url format

  #997 opened 3 years ago by xuweidongkobe
  0

• Question: Can I use cuda 11, for gpu inference on mms

  #996 opened 3 years ago by kaushal-idx
  4

• command not found: multi-model-server

  #993 opened 3 years ago by nimafo
  1

• Process run on a single CPU core

  #960 opened 4 years ago by ngoanpv
  2

• log4j2 metrics JsonLayout / QLogLayout logger broken

  #994 opened 3 years ago by lxning
  0

• `com.amazonaws.ml.mms.metrics.MetricCollector - java.io.IOException: Broken pipe` and `error while loading shared libraries: libpython3.7m.so.1.0`

  #992 opened 3 years ago by llorenzo-matterport
  0

• [Q] GPU support

  #938 opened 4 years ago by oonisim
  3

• The example in /examples/mxnet_vision/ does not work

  #991 opened 3 years ago by kylehh
  1

• Default log level change from 1.1.4 -> 1.1.6

  #987 opened 3 years ago by kastman
  4

• Upgrade log4j version to 2.17.1

  #989 opened 3 years ago by glc-froussel
  0

• Remove ineffective log4j 1 references from code

  #988 opened 3 years ago by nikhil-sk
  0

• memory utilization increment after every request, worker died, memory issue

  #974 opened 3 years ago by n0thing233
  1

• preload_model set in config.properties does not work for load model request

  #977 opened 3 years ago by lxning
  0

• Custom plug-in

  #976 opened 3 years ago by HuryanKliashchouTR
  0

• Change plugins logic

  #975 opened 3 years ago by UsernameJava
  0

• Allow custom HTTP status in mms.service.Service

  #961 opened 3 years ago by jcsaaddupuy
  3

• For multithreaded inferencing on GPU machine, with preload_model=True and default_workers_per_model=2 getting the following error

  #959 opened 4 years ago by msameedkhan
  1

• Is there anyway to yeild from MMS asynchronously?

  #971 opened 4 years ago by collinarnett
  1

• big file request will not release memory

  #967 opened 4 years ago by yangjian1218
  0

• how to build several apps and functions

  #966 opened 4 years ago by yangjian1218
  0

• SAGEMAKER_MULTI_MODE in Ping should be SAGEMAKER_MULTI_MODEL

  #963 opened 4 years ago by fm1ch4
  0

• Wrong version of ONNX as requirement of model-archiver[onnx]

  #962 opened 4 years ago by tbagrel1
  0

• Dependencies not installed in docker.

  #958 opened 4 years ago by bahar3474
  0

• define inference proto

  #951 opened 4 years ago by lxning
  1

• s3 path in tutorial is not available

  #956 opened 4 years ago by HahTK
  0

• bad links in the Model Zoo page blocking our pipelines

  #955 opened 4 years ago by junpuf
  0

• Slack link not working.

  #954 opened 4 years ago by azmaktr
  0

• tensor_gpu-inl.h:35: Check failed: e == cudaSuccess: CUDA: initialization error

  #953 opened 4 years ago by wdh234
  0

• Detect video

  #947 opened 4 years ago by wdh234
  0

• Is it possible to implement multi-stage inferences?

  #939 opened 4 years ago by bedilbek
  1

• ONNX to .MAR converter test case fails.

  #936 opened 4 years ago by quantum-fusion
  6