Skip to content

Commit aa55b40

Browse files
authored
Update how-to-save-write-experiment-files.md
1 parent 4692410 commit aa55b40

File tree

1 file changed

+3
-3
lines changed

1 file changed

+3
-3
lines changed

articles/machine-learning/service/how-to-save-write-experiment-files.md

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -40,11 +40,11 @@ Your total snapshot size exceeds the limit of 300.0 MB
4040

4141
Try one of the following solutions to resolve the error.
4242

43-
Experiment requirement|Storage limit solution
43+
Experiment requirement|Storage limit solution
4444
---|---
4545
Very large files| Move files to a datastore, and specify the datastore as your source_directory to prevent latency issues when the script folder is copied to the compute target environment at runtime.
4646
Many/ large files, but don't want a datastore| Override limit by setting SNAPSHOT_MAX_SIZE_BYTES to whatever your experiment needs. <br> `azureml._restclient.snapshots_client.SNAPSHOT_MAX_SIZE_BYTES = 'insert_desired_size'`
47-
Must use specific script directory| Make an ignore file (`.gitignore` or `.amlignore`) to prevent files from being included in the experiment snapshot that are not really a part of the source code. Place this file in the directory and add the files names to ignore to it. The `.amlignore` file uses the same syntax and patterns as the `.gitignore` file. If both files exist, the `.amlignore` file takes precedence.
47+
Must use specific script directory| Make an ignore file (`.gitignore` or `.amlignore`) to prevent files from being included in the experiment snapshot that are not really a part of the source code. Place this file in the directory and add the files names to ignore in it. The `.amlignore` file uses the same syntax and patterns as the `.gitignore` file. If both files exist, the `.amlignore` file takes precedence.
4848
Pipeline|Use a different subdirectory for each step or create an ignore file.
4949
Jupyter notebooks| You are likely using a directory that has more than 300 MB worth of data or files inside. Move your notebook into a new, empty, subdirectory with the following steps. <br> 1. Create a new folder.<br> 2. Move Jupyter notebook into empty folder. <br> 3. Run the code again.
5050

@@ -67,4 +67,4 @@ Write files to one of the following:
6767

6868
* Learn more about [accessing data from your datastores](how-to-access-data.md).
6969

70-
* Learn more about [How to Set Up Training Targets](how-to-set-up-training-targets.md).
70+
* Learn more about [How to Set Up Training Targets](how-to-set-up-training-targets.md).

0 commit comments

Comments
 (0)