For SDXL Fine-tuning

To prepare a Dataset for Fine-tuning SDXL, follow the steps below:

Custom Dataset

To prepare a dataset for fine-tuning SDXL using your own dataset, follow these steps:

💡

Upload your Dataset or provide an S3 Bucket Link:

First you will have to upload a collection of images or a zip file containing a collection of images that you want to target for finetuning SDXL models on MonsterAPI or provide an s3 bucket link if you have a collection already somewhere.

For image dataset upload, you can visit Custom Datasets Portal and upload the images as shown below:

👍

Use your uploaded Datasets or S3 bucket links:

Once you have uploaded your data, you would be able to use it as shown in the below mentioned steps:

1. Choose Data Source: Set the data source to "Custom Dataset." By default, the data source is set to "Custom Dataset," but you can adjust this based on your preference.

   

2. Select Your Dataset: Choose the dataset you uploaded in the "Datasets" section. Ensure the dataset is in a zip folder containing a collection of images.

   

Click "Next Step" and rest; Our FineTuner will take care of the rest.

---

S3 Pre-signed Link

To prepare a dataset for fine-tuning SDXL using an S3 pre-signed link, follow these steps:

1. Choose Data Source: Set the data source to "S3 Pre-signed Link."

   

2. Provide S3 URL: Enter the S3 URL of your dataset.

   

And we are done!!! Our FineTuner will take care of the rest. Simply click "Next" and finalize your fine-tuning job request.