Azure / Azure/MachineLearningNotebooks
Batch Inferencing Endpoint job run error: EntryScriptException: No module named <score file>
- 主要語言
- Jupyter Notebook
- 星號
- 4.4k
- 分支
- 2.6k
- PR 合併指標
- 30 天內沒有已合併 PR
描述
Hi, I created my Batch-Inferencing endpoint in AML studio with custom env, during deployment I uploaded my score file from my local machine, with inferencing cluster setup. Later on I tested run with some random data just to test out the pipeline. I received the EntryScriptException: No module named 'score_return_list', at the end of my job run where 'score_return_list' is the score file I uploaded during the endpoint setup process and my run() is just going to return a static list.
[tracelog.txt](https://github.com/Azure/MachineLearningNotebooks/files/8765788/tracelog.txt)
[score_return_list.txt](https://github.com/Azure/MachineLearningNotebooks/files/8765801/score_return_list.txt)
貢獻指南
這個儲存庫沒有索引到貢獻指南
研究方向
首先閱讀 tracelog.txt、score_return_list.txt,以及 issue 中描述的 endpoint 部署詳細資訊。使用已上傳的評分檔案和自訂環境重現 Batch-Inferencing endpoint 的執行,然後驗證 job 是否在沒有 EntryScriptException 的情況下完成並回傳預期的靜態清單。
由索引模型根據 Issue 內容生成。
評估
- 技術堆疊
- azure, python
- 領域
- api, cloud, machine-learning
- Issue 類型
- 缺陷
- 難度
- 4/5
- 預估耗時
- 3-5 天
- 活躍度
- 停滯
- 描述清晰度
- 需要釐清
- 新手友好度
- 25/100