prototorch_models/examples/gmlvq_iris.py

74 lines
1.8 KiB
Python
Raw Normal View History

2021-09-01 08:49:57 +00:00
"""GMLVQ example using the Iris dataset."""
import argparse
import warnings
2021-09-01 08:49:57 +00:00
import prototorch as pt
import pytorch_lightning as pl
import torch
from prototorch.models import GMLVQ, VisGMLVQ2D
from pytorch_lightning.utilities.seed import seed_everything
from pytorch_lightning.utilities.warnings import PossibleUserWarning
2021-09-01 08:49:57 +00:00
from torch.optim.lr_scheduler import ExponentialLR
from torch.utils.data import DataLoader
warnings.filterwarnings("ignore", category=PossibleUserWarning)
warnings.filterwarnings("ignore", category=UserWarning)
2021-09-01 08:49:57 +00:00
if __name__ == "__main__":
# Reproducibility
seed_everything(seed=4)
2021-09-01 08:49:57 +00:00
# Command-line arguments
parser = argparse.ArgumentParser()
parser = pl.Trainer.add_argparse_args(parser)
args = parser.parse_args()
# Dataset
train_ds = pt.datasets.Iris()
# Dataloaders
train_loader = DataLoader(train_ds, batch_size=64)
2021-09-01 08:49:57 +00:00
# Hyperparameters
hparams = dict(
input_dim=4,
latent_dim=4,
distribution={
"num_classes": 3,
"per_class": 2
},
proto_lr=0.01,
bb_lr=0.01,
)
# Initialize the model
model = GMLVQ(
2021-09-01 08:49:57 +00:00
hparams,
optimizer=torch.optim.Adam,
prototypes_initializer=pt.initializers.SMCI(train_ds),
lr_scheduler=ExponentialLR,
lr_scheduler_kwargs=dict(gamma=0.99, verbose=False),
)
# Compute intermediate input and output sizes
model.example_input_array = torch.zeros(4, 4)
# Callbacks
vis = VisGMLVQ2D(data=train_ds)
2021-09-01 08:49:57 +00:00
# Setup trainer
trainer = pl.Trainer.from_argparse_args(
args,
callbacks=[
vis,
],
max_epochs=100,
log_every_n_steps=1,
detect_anomaly=True,
2021-09-01 08:49:57 +00:00
)
# Training loop
trainer.fit(model, train_loader)