Generator

Learn how to fine-tune a model that can generate text.

Install package

pip install --upgrade tune-the-model

Set up the key to environment variable

In case you don’t have a key yet, please follow this guide.

export TTM_API_KEY=<insert your API key here>

Load the dataset

The dataset consists of lines with the following fields:

Example:

question,answer

Let’s load the dataset:

import pandas as pd


tdf = pd.read_csv('train.csv')
vdf = pd.read_csv('test.csv')

Fine tune a model

A training phase takes between 30 minutes and 5 hours depending on a dataset size.

Calling just one method will under the hood do the following for you: create a model, save it to the file, upload datasets and put the model in the queue for training.

model = ttm.train_generator(
    'generator.json',
    tdf['inputs'], tdf['outputs'],
    vdf['inputs'], vdf['outputs']
)

model.wait_for_training_finish()

Infer

the_answer = model.generate('The Answer to the Ultimate Question of Life, the Universe, and Everything')
print(the_answer)

Complete example

import tune_the_model as ttm
import pandas as pd

tdf = pd.read_csv('train.csv')
vdf = pd.read_csv('test.csv')

model = ttm.tune_generator(
    'filename.json',
    tdf['inputs'], tdf['outputs'],
    vdf['inputs'], vdf['outputs']
)

model.wait_for_training_finish()

the_answer = model.generate('The Answer to the Ultimate Question of Life, the Universe, and Everything')
print(the_answer)