test_train.py

import os

import pytest
from hydra.core.hydra_config import HydraConfig
from omegaconf import open_dict

from src.tasks import train_task
from tests.helpers.run_if import RunIf


def test_train_fast_dev_run(cfg_train):
    """Run for 1 train, val and test step."""
    HydraConfig().set_config(cfg_train)
    with open_dict(cfg_train):
        cfg_train.trainer.fast_dev_run = True
        cfg_train.trainer.accelerator = "cpu"
    train_task.train(cfg_train)


@RunIf(min_gpus=1)
def test_train_fast_dev_run_gpu(cfg_train):
    """Run for 1 train, val and test step on GPU."""
    HydraConfig().set_config(cfg_train)
    with open_dict(cfg_train):
        cfg_train.trainer.fast_dev_run = True
        cfg_train.trainer.accelerator = "gpu"
    train_task.train(cfg_train)


@RunIf(min_gpus=1)
@pytest.mark.slow
def test_train_epoch_gpu_amp(cfg_train):
    """Train 1 epoch on GPU with mixed-precision."""
    HydraConfig().set_config(cfg_train)
    with open_dict(cfg_train):
        cfg_train.trainer.max_epochs = 1
        cfg_train.trainer.accelerator = "cpu"
        cfg_train.trainer.precision = 16
    train_task.train(cfg_train)


@pytest.mark.slow
def test_train_epoch_double_val_loop(cfg_train):
    """Train 1 epoch with validation loop twice per epoch."""
    HydraConfig().set_config(cfg_train)
    with open_dict(cfg_train):
        cfg_train.trainer.max_epochs = 1
        cfg_train.trainer.val_check_interval = 0.5
    train_task.train(cfg_train)


@pytest.mark.slow
def test_train_ddp_sim(cfg_train):
    """Simulate DDP (Distributed Data Parallel) on 2 CPU processes."""
    HydraConfig().set_config(cfg_train)
    with open_dict(cfg_train):
        cfg_train.trainer.max_epochs = 2
        cfg_train.trainer.accelerator = "cpu"
        cfg_train.trainer.devices = 2
        cfg_train.trainer.strategy = "ddp_spawn"
    train_task.train(cfg_train)


@pytest.mark.slow
def test_train_resume(tmp_path, cfg_train):
    """Run 1 epoch, finish, and resume for another epoch."""
    with open_dict(cfg_train):
        cfg_train.trainer.max_epochs = 1

    HydraConfig().set_config(cfg_train)
    metric_dict_1, _ = train_task.train(cfg_train)

    files = os.listdir(tmp_path / "checkpoints")
    assert "last.ckpt" in files
    assert "epoch_000.ckpt" in files

    with open_dict(cfg_train):
        cfg_train.ckpt_path = str(tmp_path / "checkpoints" / "last.ckpt")
        cfg_train.trainer.max_epochs = 2

    metric_dict_2, _ = train_task.train(cfg_train)

    files = os.listdir(tmp_path / "checkpoints")
    assert "epoch_001.ckpt" in files
    assert "epoch_002.ckpt" not in files

    assert metric_dict_1["train/acc"] < metric_dict_2["train/acc"]
    assert metric_dict_1["val/acc"] < metric_dict_2["val/acc"]