Skip to main content
Module

x/openai/resources/mod.ts>FineTune.Hyperparams

Deno build of the official Typescript library for the OpenAI API.
openai/openai-deno-build
Extremely Popular
Go to Latest
interface FineTune.Hyperparams
import { type FineTune } from "https://deno.land/x/openai@v4.21.0/resources/mod.ts";
const { Hyperparams } = FineTune;

The hyperparameters used for the fine-tuning job. See the fine-tuning guide for more details.

Properties

batch_size: number

The batch size to use for training. The batch size is the number of training examples used to train a single forward and backward pass.

learning_rate_multiplier: number

The learning rate multiplier to use for training.

n_epochs: number

The number of epochs to train the model for. An epoch refers to one full cycle through the training dataset.

prompt_loss_weight: number

The weight to use for loss on the prompt tokens.

optional
classification_n_classes: number

The number of classes to use for computing classification metrics.

optional
classification_positive_class: string

The positive class to use for computing classification metrics.

optional
compute_classification_metrics: boolean

The classification metrics to compute using the validation dataset at the end of every epoch.