How to inject my custom datasets

[sany2k8]

On the line # 31 and # 32 of FreeDiscovery/examples/engine/duplicate_detection.py file
It says To use a custom dataset, simply specify the following variables
data_dir = input_ds['metadata']['data_dir']

Please let me know how can I inject my datasets before detecting duplicates?

I don't want to use this, I want to use my source of dataset because I want to find duplicates of my Elasticsearch documents
dataset_name = "treclegal09_2k_subset" # see list of available datasets