prototorch_models/examples/gtlvq_moons.py

80 lines
2.1 KiB
Python
Raw Normal View History

"""Localized-GTLVQ example using the Moons dataset."""
2021-11-10 17:04:24 +00:00
import argparse
import logging
import warnings
2021-11-10 17:04:24 +00:00
import prototorch as pt
import pytorch_lightning as pl
import torch
from lightning_fabric.utilities.seed import seed_everything
from prototorch.models import GTLVQ, VisGLVQ2D
from pytorch_lightning.callbacks import EarlyStopping
from pytorch_lightning.utilities.warnings import PossibleUserWarning
from torch.utils.data import DataLoader
warnings.filterwarnings("ignore", category=PossibleUserWarning)
warnings.filterwarnings("ignore", category=UserWarning)
2021-11-10 17:04:24 +00:00
if __name__ == "__main__":
# Command-line arguments
parser = argparse.ArgumentParser()
parser.add_argument("--gpus", type=int, default=0)
parser.add_argument("--fast_dev_run", type=bool, default=False)
2021-11-10 17:04:24 +00:00
args = parser.parse_args()
# Reproducibility
seed_everything(seed=2)
2021-11-10 17:04:24 +00:00
# Dataset
train_ds = pt.datasets.Moons(num_samples=300, noise=0.2, seed=42)
# Dataloaders
train_loader = DataLoader(
train_ds,
batch_size=256,
shuffle=True,
)
2021-11-10 17:04:24 +00:00
# Hyperparameters
# Latent_dim should be lower than input dim.
hparams = dict(distribution=[1, 3], input_dim=2, latent_dim=1)
2021-11-10 17:04:24 +00:00
# Initialize the model
model = GTLVQ(hparams,
prototypes_initializer=pt.initializers.SMCI(train_ds))
2021-11-10 17:04:24 +00:00
# Compute intermediate input and output sizes
model.example_input_array = torch.zeros(4, 2)
# Summary
logging.info(model)
2021-11-10 17:04:24 +00:00
# Callbacks
vis = VisGLVQ2D(data=train_ds)
es = EarlyStopping(
2021-11-10 17:04:24 +00:00
monitor="train_acc",
min_delta=0.001,
patience=20,
mode="max",
verbose=False,
check_on_train_epoch_end=True,
)
# Setup trainer
trainer = pl.Trainer(
accelerator="cuda" if args.gpus else "cpu",
devices=args.gpus if args.gpus else "auto",
fast_dev_run=args.fast_dev_run,
2021-11-10 17:04:24 +00:00
callbacks=[
vis,
es,
],
max_epochs=1000,
log_every_n_steps=1,
detect_anomaly=True,
2021-11-10 17:04:24 +00:00
)
# Training loop
trainer.fit(model, train_loader)