Steps Gallery¶
-
DeitaFiltering
Filter dataset rows using DEITA filtering strategy.
-
PushToHub
Push data to a Hugging Face Hub dataset.
-
PreferenceToArgilla
Creates a preference dataset in Argilla.
-
TextGenerationToArgilla
Creates a text generation dataset in Argilla.
-
CombineColumns
Combines columns from a list of
StepInput
. -
ExpandColumns
Expand columns that contain lists into multiple rows.
-
ConversationTemplate
Generate a conversation template from an instruction and a response.
-
FormatTextGenerationDPO
Format the output of your LLMs for Direct Preference Optimization (DPO).
-
FormatChatGenerationDPO
Format the output of a combination of a
ChatGeneration
+ a preference task such as -
FormatTextGenerationSFT
Format the output of a
TextGeneration
task for Supervised Fine-Tuning (SFT). -
FormatChatGenerationSFT
Format the output of a
ChatGeneration
task for Supervised Fine-Tuning (SFT) following the -
KeepColumns
Keeps selected columns in the dataset.
-
LoadDataFromDicts
Loads a dataset from a list of dictionaries.
-
LoadDataFromHub
Loads a dataset from the Hugging Face Hub.
-
LoadHubDataset
-
LoadDataFromFileSystem
Loads a dataset from a file in your filesystem.
-
LoadDataFromDisk
Load a dataset that was previously saved to disk.