How do I get my output images to look the same as the ones in my dataset?
There are many reasons why your images may not look the same as your dataset. These can range from issues with the size of your dataset to issues with dataset curation. You may also need to add assistive prompts. It is also possible that a different regularization class would be a better fit for what you're trying to create.
Some good general rules of thumb are to make sure that whatever dataset you are using reflects your desired outcome. If you are attempting to create a more general model, than you likely will need a larger range of very diverse images. However, if your intention is to make a generator that is capable of replicating a specific character or item, this can be achieved with a much smaller dataset. Either way, you need to make sure that none of your images are repeats of one another.
If you are having difficulty achieving your desired images, we recommend reviewing our tutorials on dataset curation and custom training generators.
Some good general rules of thumb are to make sure that whatever dataset you are using reflects your desired outcome. If you are attempting to create a more general model, than you likely will need a larger range of very diverse images. However, if your intention is to make a generator that is capable of replicating a specific character or item, this can be achieved with a much smaller dataset. Either way, you need to make sure that none of your images are repeats of one another.
If you are having difficulty achieving your desired images, we recommend reviewing our tutorials on dataset curation and custom training generators.
Updated on: 16/01/2023
Thank you!