def run(self, config: Dict[Text, Any]):
# Extract the paths to create the tar
logger.info('Orchestrating pipeline on Kubernetes..')
repo: Repository = Repository.get_instance()
repo_path = repo.path
config_dir = repo.zenml_config.config_dir
tar_file_name = \
f'{EXTRACTED_TAR_DIR_NAME}_{str(int(time.time()))}.tar.gz'
path_to_tar = os.path.join(config_dir, tar_file_name)
# Create tarfile but exclude .zenml folder if exists
path_utils.create_tarfile(repo_path, path_to_tar)
logger.info(f'Created tar of current repository at: {path_to_tar}')
# Upload tar to artifact store
store_path = config[keys.GlobalKeys.ARTIFACT_STORE]
store_staging_area = os.path.join(store_path, STAGING_AREA)
store_path_to_tar = os.path.join(store_staging_area, tar_file_name)
path_utils.copy(path_to_tar, store_path_to_tar)
logger.info(f'Copied tar to artifact store at: {store_path_to_tar}')
# Remove tar
path_utils.rm_dir(path_to_tar)
logger.info(f'Removed tar at: {path_to_tar}')
# Append path of tar in config orchestrator utils
config[keys.GlobalKeys.BACKEND][
keys.BackendKeys.ARGS][TAR_PATH_ARG] = store_path_to_tar
# Launch the instance
self.launch_job(config)
def run_fn(self):
train_dataset = self.input_fn(self.train_files,
self.tf_transform_output)
eval_dataset = self.input_fn(self.eval_files, self.tf_transform_output)
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
model = self.model_fn(train_dataset, eval_dataset)
model.to(device)
criterion = nn.BCEWithLogitsLoss()
optimizer = optim.Adam(model.parameters(), lr=0.001)
model.train()
for e in range(1, self.epoch + 1):
epoch_loss = 0
epoch_acc = 0
step_count = 0
for x, y in train_dataset:
step_count += 1
X_batch, y_batch = x.to(device), y.to(device)
optimizer.zero_grad()
y_pred = model(X_batch)
loss = criterion(y_pred, y_batch)
acc = binary_acc(y_pred, y_batch)
loss.backward()
optimizer.step()
epoch_loss += loss.item()
epoch_acc += acc.item()
print(f'Epoch {e + 0:03}: | Loss: '
f'{epoch_loss / step_count:.5f} | Acc: '
f'{epoch_acc / step_count:.3f}')
path_utils.create_dir_if_not_exists(self.serving_model_dir)
if path_utils.is_remote(self.serving_model_dir):
temp_model_dir = '__temp_model_dir__'
temp_path = os.path.join(os.getcwd(), temp_model_dir)
if path_utils.is_dir(temp_path):
raise PermissionError('{} is used as a temp path but it '
'already exists. Please remove it to '
'continue.')
torch.save(model, temp_path)
path_utils.copy_dir(temp_path, self.serving_model_dir)
path_utils.rm_dir(temp_path)
else:
torch.save(model, os.path.join(self.serving_model_dir, 'model.pt'))
def run(self, config: Dict[Text, Any]):
"""
This run function essentially calls an underlying TFX orchestrator run.
However it is meant as a higher level abstraction with some
opinionated decisions taken.
Args:
config: a ZenML config dict
"""
# Extract the paths to create the tar
logger.info('Orchestrating pipeline on GCP..')
repo: Repository = Repository.get_instance()
repo_path = repo.path
config_dir = repo.zenml_config.config_dir
tar_file_name = \
f'{EXTRACTED_TAR_DIR_NAME}_{str(int(time.time()))}.tar.gz'
path_to_tar = os.path.join(config_dir, tar_file_name)
# Create tarfile but excluse .zenml folder if exists
path_utils.create_tarfile(repo_path, path_to_tar)
logger.info(f'Created tar of current repository at: {path_to_tar}')
# Upload tar to artifact store
store_path = config[keys.GlobalKeys.ARTIFACT_STORE]
store_staging_area = os.path.join(store_path, STAGING_AREA)
store_path_to_tar = os.path.join(store_staging_area, tar_file_name)
path_utils.copy(path_to_tar, store_path_to_tar)
logger.info(f'Copied tar to artifact store at: {store_path_to_tar}')
# Remove tar
path_utils.rm_dir(path_to_tar)
logger.info(f'Removed tar at: {path_to_tar}')
# Append path of tar in config orchestrator utils
config[keys.GlobalKeys.BACKEND][
keys.BackendKeys.ARGS][TAR_PATH_ARG] = store_path_to_tar
# Launch the instance
self.launch_instance(config)
logger = get_logger(__name__)
# reset pipeline root to redirect to tests so that it writes the yamls there
ZENML_ROOT = str(Path(zenml.__path__[0]).parent)
TEST_ROOT = os.path.join(ZENML_ROOT, "tests")
Repository.init_repo(TEST_ROOT, analytics_opt_in=False)
pipeline_root = os.path.join(TEST_ROOT, "pipelines")
csv_root = os.path.join(TEST_ROOT, "test_data")
image_root = os.path.join(csv_root, "images")
repo: Repository = Repository.get_instance()
if path_utils.is_dir(pipeline_root):
path_utils.rm_dir(pipeline_root)
repo.zenml_config.set_pipelines_dir(pipeline_root)
try:
for i in range(1, 6):
training_pipeline = TrainingPipeline(name='csvtest{0}'.format(i))
try:
# Add a datasource. This will automatically track and version it.
ds = CSVDatasource(name='my_csv_datasource',
path=os.path.join(csv_root, "my_dataframe.csv"))
except AlreadyExistsException:
ds = repo.get_datasource_by_name("my_csv_datasource")
training_pipeline.add_datasource(ds)
def run_fn(self):
train_dataset = self.input_fn(self.train_files,
self.tf_transform_output)
eval_dataset = self.input_fn(self.eval_files, self.tf_transform_output)
class LitModel(pl.LightningModule):
def __init__(self):
super().__init__()
self.l1 = torch.nn.Linear(8, 64)
self.layer_out = torch.nn.Linear(64, 1)
def forward(self, x):
x = torch.relu(self.l1(x))
x = self.layer_out(x)
return x
def training_step(self, batch, batch_idx):
x, y = batch
y_hat = self(x)
loss = F.binary_cross_entropy_with_logits(y_hat, y)
tensorboard_logs = {'train_loss': loss}
return {'loss': loss, 'log': tensorboard_logs}
def configure_optimizers(self):
return torch.optim.Adam(self.parameters(), lr=0.001)
def train_dataloader(self):
return train_dataset
def validation_step(self, batch, batch_idx):
x, y = batch
y_hat = self(x)
return {
'val_loss': F.binary_cross_entropy_with_logits(y_hat, y)
}
def validation_epoch_end(self, outputs):
avg_loss = torch.stack([x['val_loss'] for x in outputs]).mean()
tensorboard_logs = {'val_loss': avg_loss}
return {'avg_val_loss': avg_loss, 'log': tensorboard_logs}
def val_dataloader(self):
return eval_dataset
model = LitModel()
# most basic trainer, uses good defaults
trainer = Trainer(
default_root_dir=self.log_dir,
max_epochs=self.epoch,
)
trainer.fit(model)
path_utils.create_dir_if_not_exists(self.serving_model_dir)
if path_utils.is_remote(self.serving_model_dir):
temp_model_dir = '__temp_model_dir__'
temp_path = os.path.join(os.getcwd(), temp_model_dir)
if path_utils.is_dir(temp_path):
raise PermissionError('{} is used as a temp path but it '
'already exists. Please remove it to '
'continue.')
trainer.save_checkpoint(os.path.join(temp_path, 'model.cpkt'))
path_utils.copy_dir(temp_path, self.serving_model_dir)
path_utils.rm_dir(temp_path)
else:
trainer.save_checkpoint(
os.path.join(self.serving_model_dir, 'model.ckpt'))
def run_fn(self):
train_dataset = self.input_fn(self.train_files,
self.tf_transform_output)
eval_dataset = self.input_fn(self.eval_files,
self.tf_transform_output)
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
model = self.model_fn(train_dataset, eval_dataset)
model.to(device)
criterion = nn.BCEWithLogitsLoss()
optimizer = optim.Adam(model.parameters(), lr=0.001)
writer = SummaryWriter(self.log_dir)
model.train()
total_count = 0
for e in range(1, self.epochs + 1):
epoch_loss = 0
epoch_acc = 0
step_count = 0
for x, y, _ in train_dataset:
step_count += 1
total_count += 1
x_batch = torch.cat([v.to(device) for v in x.values()], dim=-1)
y_batch = torch.cat([v.to(device) for v in y.values()], dim=-1)
optimizer.zero_grad()
y_pred = model(x_batch)
loss = criterion(y_pred, y_batch)
acc = binary_acc(y_pred, y_batch)
loss.backward()
optimizer.step()
epoch_loss += loss.item()
epoch_acc += acc.item()
if e == 1 and step_count == 1:
writer.add_graph(model, x_batch)
writer.add_scalar('training_loss', loss, total_count)
writer.add_scalar('training_accuracy', acc, total_count)
print(f'Epoch {e + 0:03}: | Loss: '
f'{epoch_loss / step_count:.5f} | Acc: '
f'{epoch_acc / step_count:.3f}')
# test
test_results = self.test_fn(model, eval_dataset)
utils.save_test_results(test_results, self.test_results)
path_utils.create_dir_if_not_exists(self.serving_model_dir)
if path_utils.is_remote(self.serving_model_dir):
temp_model_dir = '__temp_model_dir__'
temp_path = os.path.join(os.getcwd(), temp_model_dir)
if path_utils.is_dir(temp_path):
raise PermissionError('{} is used as a temp path but it '
'already exists. Please remove it to '
'continue.')
torch.save(model, temp_path)
path_utils.copy_dir(temp_path, self.serving_model_dir)
path_utils.rm_dir(temp_path)
else:
torch.save(model, os.path.join(self.serving_model_dir, 'model.pt'))
