Text Generation
Transformers
PyTorch
TensorBoard
Safetensors
bloom
Eval Results
text-generation-inference
Inference Endpoints

How to use sampling in Interference API

#136
by imwide - opened

This is my code:

def query(payload):
    response = requests.post(API_URL, headers=headers, json=payload)
    return response.json()

output = query({"inputs": promt, "do_sample": True})

Strangeley, adding the "do sample" part in the query's arguments doesnt change anything. The answers are always the same with the same promt too, but I want some variation. Anyone know what Im doing wrong?

BigScience Workshop org

Duplicate of https://huggingface.co/bigscience/bloom/discussions/131#6368094b5bb06007ea0d5b3c . Let's close this one in favor of the other discussion.

TimeRobber changed discussion status to closed

Sign up or log in to comment