Migrate From DLS Desktop 2.x version to Latest DLS 3.x
This guide is for users to show how to migrate Deep Learning Studio (DLS) Desktop 2.x version to the latest 3.0 version.
To check if you are still using the older DLS Desktop version, please check the version number on the bottom right corner after launching DLS:

1) Creation of Backups from DLS Desktop 2.x

2) Restoration of Backups on DLS 3.x (Desktop/Cloud)

1 (a) Creation of Datasets Backup from DLS Desktop 2.x

    Create a backup of dataset folder from the path
      <DLS_FOLDER>/data/1/datasets
      Default location : - “C:/Users/<username>/AppData/Local/Programs/DeepLearningStudio/data/1/dataset”
    Create a backup of db.sqlite3 file from the path
      “<DLS_FOLDER>/home/app/database/db.sqlite3”
      Default location : -
      C:/Users/<username>/AppData/Local/Programs/DeepLearningStudio/home/app/database/db.sqlite3"

1 (b) Creation of Projects Backup from DLS Desktop 2.x

    Open the project and go to the Model tab, and download the model config by clicking on Download icon
Important
Repeat the above steps for every project to create a backup

2 (a) Restoration of Backups on DLS Desktop 3.x (Windows / Ubuntu)

Important
Make sure you have installed DLS 3.0 Desktop Version. If not then refer to Windows and Ubuntu installation guide
Restore datasets:
    1.
    Open dataset backup folder
    2.
    Download this file and place it inside all the datasets folder
    3.
    Copy the dataset folder and paste it to <DLS_FOLDER>/user_data/data/1/datasets/
Restore projects:
    1.
    Restore the db file by moving it to DLS 3.0 <DLS_FOLDER>/user_data/data/database/ copy the “db.sqlite3” file and paste it to the DLS new version at <DLS_FOLDER>/user_data/data/database/
    2.
    Start DLS
    3.
    Open or Create your projects
    4.
    Import the downloaded model config files (.yaml )

2 (b) Restoration of Backups on DLS Cloud 3.x (AWS / Azure)

Important
Make sure you have launched DLS in AWS/Azure Cloud, if not please launch or you can also refer Azure and AWS installation guide
Restore datasets:
    1.
    Open dataset backup folder
    2.
    You need to zip each dataset separately and upload it as DLS Native format from dataset menu, ref. link
Restore projects:
    1.
    Start DLS
    2.
    Create your projects
    3.
    Import the downloaded model config files (.yaml )
Notes :
    As we have upgraded to a new version, you may face some conflicts in your project model.
    Refer to this note for the arguments/parameters mismatch in the model layers.
Congratulations !!
You have successfully retained your data and ready to begin with, our upgraded version.
Contact us for any further assistance. We would be happy to help you.
Last modified 1yr ago