Azure / Azure/MachineLearningNotebooks
Clarity on downloading labelled dataset
- 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 used this code from here:
```
from azureml.core import Workspace, Dataset
subscription_id = ''
resource_group = ''
workspace_name = ''
workspace = Workspace(subscription_id, resource_group, workspace_name)
dataset = Dataset.get_by_name(workspace, name='')
dataset.download(target_path='.', overwrite=False)
```
https://ml.azure.com/dataset/NewSalmonFrames/1/details?wsid=/subscriptions/4e93dac0-d1e9-41a5-8e98-9707b26b6913/resourcegroups/eli_kling_speacies/providers/Microsoft.MachineLearningServices/workspaces/eli_kling_speacies&tid=62368045-c91f-477d-a117-8432f56a91e4#consumeTab
This works created, downloading the labelled dataset in the COCO format. What if i wanted to download it in the AzureML format, or using aboslute bounding box co-ordinates?
I can manually export from the UI, but is there a way to do it programatically. I'm worried Azure will change this feature in the future, and my re-usable code will not be so re-useable.
Thanls,
Viraj
貢獻指南
這個儲存庫沒有索引到貢獻指南
研究方向
從所示的 Azure ML Python SDK 入口點開始:Dataset.get_by_name 和 Dataset.download,然後將它們與連結的資料集 UI 的匯出選項進行比較。確認是否以程式方式支援 AzureML 格式或絕對邊界框座標;完成的標準是記錄受支援的 API,或確認只有 UI 匯出可用。
由索引模型根據 Issue 內容生成。
評估
- 技術堆疊
- azure, machine-learning, python
- 領域
- data, machine-learning
- Issue 類型
- 功能
- 難度
- 4/5
- 預估耗時
- 3-5 天
- 活躍度
- 停滯
- 描述清晰度
- 基本清楚
- 新手友好度
- 25/100