开发者社区 > 大数据与机器学习 > 人工智能平台PAI > 正文

使用机器学习PAI 运行训练,报了下面的问题,是怎么回事呢?

使用机器学习PAI easyrec 中的negative_sampler_v2 ,传入user table, user-pos item table 和 item table, 运行训练,报了下面的问题,是怎么回事呢?

[2024-07-26 15:02:43.233635] [ERROR] [67#469] [tensorflow/contrib/star/star_client_tag.cc:141] wrapper_done, status not ok. status code=10, err msg=Easy connection closed by remote peer.
[[{{node item_tower/embedding_lookup_7_S5323}} = _Recvclient_terminated=false, recv_device="/job:worker/replica:0/task:7/device:CPU:0", send_device="/job:ps/replica:0/task:5/device:CPU:0", send_device_incarnation=3438114829932328181, tensor_name="edge_1337_item_tower/embedding_lookup_7", tensor_type=DT_FLOAT, _device="/job:worker/replica:0/task:7/device:CPU:0"]]
[2024-07-26 15:02:43.234336] [ERROR] [67#469] [tensorflow/contrib/star/star_client_tag.cc:141] wrapper_done, status not ok. status code=10, err msg=Easy connection closed by remote peer.
[[{{node item_tower/embedding_lookup_7_S5323}} = _Recvclient_terminated=false, recv_device="/job:worker/replica:0/task:15/device:CPU:0", send_device="/job:ps/replica:0/task:5/device:CPU:0", send_device_incarnation=3438114829932328181, tensor_name="edge_1337_item_tower/embedding_lookup_7", tensor_type=DT_FLOAT, _device="/job:worker/replica:0/task:15/device:CPU:0"]]
[2024-07-26 15:02:43.235410] [ERROR] [67#469] [tensorflow/contrib/star/star_client_tag.cc:141] wrapper_done, status not ok. status code=10, err msg=Easy connection closed by remote peer.
[[{{node item_tower/embedding_lookup_7_S5323}} = _Recvclient_terminated=false, recv_device="/job:worker/replica:0/task:15/device:CPU:0", send_device="/job:ps/replica:0/task:5/device:CPU:0", send_device_incarnation=3438114829932328181, tensor_name="edge_1337_item_tower/embedding_lookup_7", tensor_type=DT_FLOAT, _device="/job:worker/replica:0/task:15/device:CPU:0"]]
[2024-07-26 15:02:43.236392] [ERROR] [67#469] [tensorflow/contrib/star/star_client_tag.cc:141] wrapper_done, status not ok. status code=14, err msg=easy conn fail/timeout

展开
收起
真的很搞笑 2024-07-31 12:15:43 47 0
0 条回答
写回答
取消 提交回答

人工智能平台 PAI(Platform for AI,原机器学习平台PAI)是面向开发者和企业的机器学习/深度学习工程平台,提供包含数据标注、模型构建、模型训练、模型部署、推理优化在内的AI开发全链路服务,内置140+种优化算法,具备丰富的行业场景插件,为用户提供低门槛、高性能的云原生AI工程化能力。

相关产品

  • 人工智能平台 PAI
  • 热门讨论

    热门文章

    相关电子书

    更多
    大规模机器学习在蚂蚁+阿里的应用 立即下载
    阿里巴巴机器学习平台AI 立即下载
    机器学习及人机交互实战 立即下载