Large Data Files
If you are tuning on a large file of data, you can use the upload_file function to first upload the file onto the servers.
Here is an example with a test.csv file:
// code/test.csv
user,answer
"Explain the process of photosynthesis","Photosynthesis is the process by which plants and some other organisms convert light energy into chemical energy. It is critical for the existence of the vast majority of life on Earth. It is the way in which virtually all energy in the biosphere becomes available to living things.
"What is the capital of USA?", "Washington, D.C."
You can use the Lamini to tune on this file directly by uploading the file and specifying the input and output keys.
# code/large_data_files_csv.py
from lamini import Lamini
llm = Lamini(model_name="meta-llama/Meta-Llama-3.1-8B-Instruct")
dataset_id = llm.upload_file("test.csv", input_key="user", output_key="answer")
llm.tune(data_or_dataset_id=dataset_id)
Alternatively, you can also use jsonlines files
Using test.jsonl
// code/test.jsonl
{"user": "Explain the process of photosynthesis", "answer": "Photosynthesis is the process by which plants and some other organisms convert light energy into chemical energy. It is critical for the existence of the vast majority of life on Earth. It is the way in which virtually all energy in the biosphere becomes available to living things."}
{"user": "What is the capital of USA?", "answer": "Washington, D.C."}