From d149c7f22fc56aa0a176b0b5bbdcabb75d1b88aa Mon Sep 17 00:00:00 2001 From: MRXLT Date: Wed, 29 Jul 2020 11:33:38 +0800 Subject: [PATCH] add FAQ --- doc/FAQ.md | 5 ++++- python/paddle_serving_server_gpu/__init__.py | 2 +- 2 files changed, 5 insertions(+), 2 deletions(-) diff --git a/doc/FAQ.md b/doc/FAQ.md index 3bdd2dfd..bb567b78 100644 --- a/doc/FAQ.md +++ b/doc/FAQ.md @@ -12,4 +12,7 @@ client.load_client_config(sys.argv[1]) client.set_rpc_timeout_ms(100000) client.connect(["127.0.0.1:9393"]) - ``` + ``` + +- Q: 执行GPU预测时出现显存不足的问题应该怎么办？ + A: 请使用--thread 4参数限制server端的线程数为4，并使用--mem_optim参数开启显存优化选项。 diff --git a/python/paddle_serving_server_gpu/__init__.py b/python/paddle_serving_server_gpu/__init__.py index 0261003a..ee926221 100644 --- a/python/paddle_serving_server_gpu/__init__.py +++ b/python/paddle_serving_server_gpu/__init__.py @@ -41,7 +41,7 @@ from concurrent import futures def serve_args(): parser = argparse.ArgumentParser("serve") parser.add_argument( - "--thread", type=int, default=10, help="Concurrency of server") + "--thread", type=int, default=4, help="Concurrency of server") parser.add_argument( "--model", type=str, default="", help="Model for serving") parser.add_argument( -- GitLab