You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: articles/machine-learning/how-to-track-monitor-analyze-runs.md
+49-59Lines changed: 49 additions & 59 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -8,38 +8,35 @@ ms.subservice: mlops
8
8
ms.author: amipatel
9
9
author: amibp
10
10
ms.reviewer: sgilley
11
-
ms.date: 06/24/2022
11
+
ms.date: 03/22/2024
12
12
ms.topic: how-to
13
13
ms.custom: build-2023
14
14
---
15
15
16
16
# Monitor and analyze jobs in studio
17
17
18
+
You can use [Azure Machine Learning studio](https://ml.azure.com) to monitor, organize, and track your jobs for training and experimentation. Your machine learning (ML) job history is an important part of an explainable and repeatable ML development process.
18
19
19
-
You can use [Azure Machine Learning studio](https://ml.azure.com) to monitor, organize, and track your jobs for training and experimentation. Your ML job history is an important part of an explainable and repeatable ML development process.
20
+
This article explains how to:
20
21
21
-
This article shows how to do the following tasks:
22
-
23
-
* Add job display name.
24
-
* Create a custom view.
25
-
* Add a job description.
22
+
* Add a job display name.
23
+
* Create a custom view.
24
+
* Add a job description.
26
25
* Tag and find jobs.
27
26
* Run search over your job history.
28
27
* Cancel or fail jobs.
29
-
* Monitor the job status by email notification.
30
-
* Monitor your job resources (preview)
28
+
* Monitor job status by email notification.
29
+
* Monitor your job resources (preview).
31
30
32
-
33
31
> [!TIP]
34
-
> * If you're looking for information on using the Azure Machine Learning SDK v1 or CLI v1, see [How to track, monitor, and analyze jobs (v1)](./v1/how-to-track-monitor-analyze-runs.md).
35
-
> * If you're looking for information on monitoring training jobs from the CLI or SDK v2, see [Track experiments with MLflow and CLI v2](how-to-use-mlflow-cli-runs.md).
36
-
> * If you're looking for information on monitoring the Azure Machine Learning service and associated Azure services, see [How to monitor Azure Machine Learning](monitor-azure-machine-learning.md).
37
-
>
38
-
> If you're looking for information on monitoring models deployed to online endpoints, see [Monitor online endpoints](how-to-monitor-online-endpoints.md).
32
+
> * For information on using the Azure Machine Learning SDK v1 or CLI v1, see [Monitor and analyze jobs in studio (v1)](./v1/how-to-track-monitor-analyze-runs.md).
33
+
> * To learn how to monitor training jobs from the CLI or SDK v2, see [Track ML experiments and models with MLflow (v2)](how-to-use-mlflow-cli-runs.md).
34
+
> * To learn how to monitor the Azure Machine Learning service and associated Azure services, see [Monitor Azure Machine Learning](monitor-azure-machine-learning.md).
35
+
> * To learn how to monitor models deployed to online endpoints, see [Monitor online endpoints](how-to-monitor-online-endpoints.md).
39
36
40
37
## Prerequisites
41
38
42
-
You'll need the following items:
39
+
You need the following items:
43
40
44
41
* To use Azure Machine Learning, you must have an Azure subscription. If you don't have an Azure subscription, create a free account before you begin. Try the [free or paid version of Azure Machine Learning](https://azure.microsoft.com/free/).
45
42
* You must have an Azure Machine Learning workspace. A workspace is created in [Install, set up, and use the CLI (v2)](how-to-configure-cli.md).
@@ -58,90 +55,83 @@ The job display name is an optional and customizable name that you can provide f
58
55
59
56
:::image type="content" source="media/how-to-track-monitor-analyze-runs/display-name.gif" alt-text="Screenshot of how to edit the display name.":::
60
57
61
-
## Custom View
62
-
58
+
## Custom view
59
+
63
60
To view your jobs in the studio:
64
61
65
62
1. Navigate to the **Jobs** tab.
66
63
67
-
1. Select either **All experiments** to view all the jobs in an experiment or select **All jobs** to view all the jobs submitted in the Workspace.
68
-
69
-
In the **All jobs'** page, you can filter the jobs list by tags, experiments, compute target and more to better organize and scope your work.
70
-
71
-
1. Make customizations to the page by selecting jobs to compare, adding charts or applying filters. These changes can be saved as a **Custom View** so you can easily return to your work. Users with workspace permissions can edit, or view the custom view. Also, share the custom view with team members for enhanced collaboration by selecting **Share view**.
64
+
1. Select either **All experiments** to view all the jobs in an experiment, or select **All jobs** to view all the jobs submitted in the Workspace.
72
65
73
-
1. To view the job logs, select a specific job and in the **Outputs + logs** tab, you can find diagnostic and error logs for your job.
66
+
On the **All jobs** page, you can filter the jobs list by tags, experiments, compute target, and more to better organize and scope your work.
67
+
68
+
1. Make customizations to the page by selecting jobs to compare, adding charts, or applying filters. These changes can be saved as a **Custom view** so you can easily return to your work. Users with workspace permissions can edit or view the custom view. Also, share the custom view with team members for enhanced collaboration by selecting **Share view**.
74
69
75
70
:::image type="content" source="media/how-to-track-monitor-analyze-runs/custom-views-2.gif" alt-text="Screenshot of how to create a custom view.":::
76
71
72
+
1. To view the job logs, select a specific job and in the **Outputs + logs** tab, you can find diagnostic and error logs for your job.
73
+
77
74
## Job description
78
75
79
-
A job description can be added to a job to provide more context and information to the job. You can also search on these descriptions from the jobs list and add the job description as a column in the jobs list.
76
+
You can add a job description to provide more context and information to the job. You can also search on these descriptions from the jobs list and add the job description as a column in the jobs list.
80
77
81
-
Navigate to the **Job Details**page for your job and select the edit or pencil icon to add, edit, or delete descriptions for your job. To persist the changes to the jobs list, save the changes to your existing Custom View or a new Custom View. Markdown format is supported for job descriptions, which allows images to be embedded and deep linking as shown below.
78
+
Navigate to the details page for your job and select the edit or pencil icon to add, edit, or delete descriptions for your job. To persist the changes to the jobs list, save the changes to your existing custom view or a new custom view. Markdown format is supported for job descriptions, which allows images to be embedded and deep linking as shown.
82
79
83
80
:::image type="content" source="media/how-to-track-monitor-analyze-runs/run-description-2.gif" alt-text="Screenshot of how to create a job description.":::
84
81
85
-
86
82
## Tag and find jobs
87
83
88
84
In Azure Machine Learning, you can use properties and tags to help organize and query your jobs for important information.
89
85
90
-
* Edit tags
86
+
###Edit tags
91
87
92
-
You can add, edit, or delete job tags from the studio. Navigate to the **Job Details** page for your job and select the edit, or pencil icon to add, edit, or delete tags for your jobs. You can also search and filter on these tags from the jobs list page.
93
-
94
-
:::image type="content" source="media/how-to-track-monitor-analyze-runs/run-tags.gif" alt-text="Screenshot of how to add, edit, or delete job tags.":::
95
-
88
+
You can add, edit, or delete job tags from the studio. Navigate to the details page for your job and select the edit or pencil icon to add, edit, or delete tags for your jobs. You can also search and filter on these tags from the jobs list page.
96
89
97
-
* Query properties and tags
90
+
:::image type="content" source="media/how-to-track-monitor-analyze-runs/run-tags.gif" alt-text="Screenshot of how to add, edit, or delete job tags.":::
98
91
99
-
You can query jobs within an experiment to return a list of jobs that match specific properties and tags.
100
-
101
-
To search for specific jobs, navigate to the **All jobs** list. From there you have two options:
102
-
103
-
1. Use the **Add filter**button and select filter on tags to filter your jobs by tag that was assigned to the job(s). <br><br>
104
-
OR
105
-
106
-
1.Use the search bar to quickly find jobs by searching on the job metadata like the job status, descriptions, experiment names, and submitter name.
92
+
### Query properties and tags
93
+
94
+
You can query jobs within an experiment to return a list of jobs that match specific properties and tags.
95
+
96
+
To search for specific jobs, navigate to the **All jobs**list. From there, you have two options:
97
+
98
+
- Use the **Add filter** button and select filter on tags to filter your jobs by tag that was assigned to the job(s).
99
+
-Use the search bar to quickly find jobs by searching on the job metadata like job status, descriptions, experiment names, and submitter name.
107
100
108
101
## Cancel or fail jobs
109
102
110
103
If you notice a mistake or if your job is taking too long to finish, you can cancel the job.
111
104
112
-
To cancel a job in the studio, using the following steps:
105
+
To cancel a job in the studio:
113
106
114
107
1. Go to the running pipeline in either the **Jobs** or **Pipelines** section.
115
108
116
109
1. Select the pipeline job number you want to cancel.
117
110
118
111
1. In the toolbar, select **Cancel**.
119
112
120
-
## Monitor the job status by email notification
113
+
## Monitor job status by email notification
121
114
122
-
1. In the [Azure portal](https://portal.azure.com/), in the left navigation bar, select the **Monitor** tab.
115
+
1. In the [Azure portal](https://portal.azure.com), in the left navigation bar, select the **Monitor** tab.
123
116
124
-
1. Select **Diagnostic settings** and then select**+ Add diagnostic setting**.
117
+
1. Select **Diagnostic settings**, then choose**+ Add diagnostic setting**.
125
118
126
-
119
+
:::image type="content" source="media/how-to-track-monitor-analyze-runs/diagnostic-setting.png" alt-text="Screenshot of diagnostic settings for email notification.":::
127
120
128
-
1. In the Diagnostic Setting,
129
-
1. under the **Category details**, select the **AmlRunStatusChangedEvent**.
130
-
1. In the **Destination details**, select the **Send to Log Analytics workspace** and specify the **Subscription** and **Log Analytics workspace**.
121
+
1. Under **Category details**, select **AmlRunStatusChangedEvent**. Under **Destination details**, select **Send to Log Analytics workspace** and specify the **Subscription** and **Log Analytics workspace**.
131
122
132
-
> [!NOTE]
133
-
> The **Azure Log Analytics Workspace** is a different type of Azure Resource than the **Azure Machine Learning service Workspace**. If there are no options in that list, you can [create a Log Analytics Workspace](../azure-monitor/logs/quick-create-workspace.md).
134
-
135
-
123
+
:::image type="content" source="media/how-to-track-monitor-analyze-runs/log-location.png" alt-text="Screenshot of where to save email notification.":::
136
124
137
-
1. In the **Logs** tab, add a **New alert rule**.
125
+
> [!NOTE]
126
+
> The **Azure Log Analytics Workspace** is a different type of Azure resource than the **Azure Machine Learning service workspace**. If there are no options in that list, you can [create a Log Analytics workspace](../azure-monitor/logs/quick-create-workspace.md).
1. In the **Logs** tab, select **New alert rule**.
140
129
141
-
1. See [how to create and manage log alerts using Azure Monitor](../azure-monitor/alerts/alerts-log.md).
130
+
:::image type="content" source="media/how-to-track-monitor-analyze-runs/new-alert-rule.png" alt-text="Screenshot of button to add new alert rule.":::
142
131
132
+
1. To learn how to create and manage log alerts using Azure Monitor, see [Create or edit a log search alert rule](../azure-monitor/alerts/alerts-log.md).
143
133
144
-
## Next steps
134
+
## Related content
145
135
146
-
*To learn how to log metrics for your experiments, see [Log metrics during training jobs](how-to-log-view-metrics.md).
147
-
*To learn how to monitor resources and logs from Azure Machine Learning, see [Monitoring Azure Machine Learning](monitor-azure-machine-learning.md).
136
+
*[Log metrics, parameters, and files with MLflow](how-to-log-view-metrics.md)
0 commit comments