def test_hook_shap(tmpdir):
np.random.seed(42)
train_data = np.random.rand(10, 10)
train_label = np.random.randint(2, size=10)
dtrain = xgboost.DMatrix(train_data, label=train_label)
out_dir = os.path.join(tmpdir, str(uuid.uuid4()))
hook = Hook(out_dir=out_dir,
include_collections=["average_shap", "full_shap"],
train_data=dtrain)
run_xgboost_model(hook=hook)
trial = create_trial(out_dir)
tensors = trial.tensor_names()
assert len(tensors) > 0
assert "average_shap" in trial.collections()
assert "full_shap" in trial.collections()
assert any(t.startswith("average_shap/") for t in tensors)
assert any(t.startswith("full_shap/") for t in tensors)
assert not any(t.endswith("/bias") for t in tensors)
average_shap_tensors = [
t for t in tensors if t.startswith("average_shap/")
]
average_shap_tensor_name = average_shap_tensors.pop()
assert trial.tensor(average_shap_tensor_name).value(0).shape == (1, )
full_shap_tensors = [t for t in tensors if t.startswith("full_shap/")]
full_shap_tensor_name = full_shap_tensors.pop()
# full shap values should have 10 rows with 10 features + 1 bias
assert trial.tensor(full_shap_tensor_name).value(0).shape == (10, 11)
def test_hook_validation(tmpdir):
np.random.seed(42)
train_data = np.random.rand(5, 10)
train_label = np.random.randint(2, size=5)
dtrain = xgboost.DMatrix(train_data, label=train_label)
valid_data = np.random.rand(5, 10)
valid_label = np.random.randint(2, size=5)
dvalid = xgboost.DMatrix(valid_data, label=valid_label)
out_dir = os.path.join(tmpdir, str(uuid.uuid4()))
hook = Hook(
out_dir=out_dir,
include_collections=["labels", "predictions"],
train_data=dtrain,
validation_data=dvalid,
)
run_xgboost_model(hook=hook)
trial = create_trial(out_dir)
tensors = trial.tensor_names()
assert len(tensors) > 0
assert "labels" in trial.collections()
assert "predictions" in trial.collections()
assert "labels" in tensors
assert "predictions" in tensors
def test_hook_save_every_step(tmpdir):
save_config = SaveConfig(save_interval=1)
out_dir = os.path.join(tmpdir, str(uuid.uuid4()))
hook = Hook(out_dir=out_dir, save_config=save_config)
run_xgboost_model(hook=hook)
trial = create_trial(out_dir)
assert trial.steps() == list(range(10))
def test_hook_from_json_config_for_losses(tmpdir, monkeypatch, params):
out_dir = tmpdir.join("test_hook_from_json_config_for_losses")
config_file = tmpdir.join("config.json")
config_file.write(get_json_config_for_losses(str(out_dir)))
monkeypatch.setenv(CONFIG_FILE_PATH_ENV_STR, str(config_file))
hook = Hook.create_from_json_file()
assert has_training_ended(out_dir) is False
run_xgboost_model(hook=hook, params=params)
trial = create_trial(str(out_dir))
eval_metric = params["eval_metric"]
test_metric = f"test-{eval_metric}"
train_metric = f"train-{eval_metric}"
if eval_metric == "rmse":
assert train_metric in trial.tensor_names(
collection=CollectionKeys.METRICS)
assert train_metric in trial.tensor_names(
collection=CollectionKeys.LOSSES)
assert test_metric in trial.tensor_names(
collection=CollectionKeys.METRICS)
assert test_metric in trial.tensor_names(
collection=CollectionKeys.LOSSES)
if eval_metric == "auc" or eval_metric == "map":
assert train_metric in trial.tensor_names(
collection=CollectionKeys.METRICS)
assert train_metric not in trial.tensor_names(
collection=CollectionKeys.LOSSES)
assert test_metric in trial.tensor_names(
collection=CollectionKeys.METRICS)
assert test_metric not in trial.tensor_names(
collection=CollectionKeys.LOSSES)
def test_hook_from_json_config_full(tmpdir, monkeypatch):
out_dir = tmpdir.join("test_hook_from_json_config_full")
config_file = tmpdir.join("config.json")
config_file.write(get_json_config_full(str(out_dir)))
monkeypatch.setenv(CONFIG_FILE_PATH_ENV_STR, str(config_file))
hook = Hook.create_from_json_file()
assert has_training_ended(out_dir) is False
run_xgboost_model(hook=hook)
def main():
args = parse_args()
train_files_path, validation_files_path = args.train, args.validation
train_features_path = os.path.join(args.train, 'train_features.csv')
train_labels_path = os.path.join(args.train, 'train_labels.csv')
val_features_path = os.path.join(args.validation, 'val_features.csv')
val_labels_path = os.path.join(args.validation, 'val_labels.csv')
print('Loading training dataframes...')
df_train_features = pd.read_csv(train_features_path, header=None)
df_train_labels = pd.read_csv(train_labels_path, header=None)
print('Loading validation dataframes...')
df_val_features = pd.read_csv(val_features_path, header=None)
df_val_labels = pd.read_csv(val_labels_path, header=None)
X = df_train_features.values
y = df_train_labels.values.reshape(-1)
val_X = df_val_features.values
val_y = df_val_labels.values.reshape(-1)
print('Train features shape: {}'.format(X.shape))
print('Train labels shape: {}'.format(y.shape))
print('Validation features shape: {}'.format(val_X.shape))
print('Validation labels shape: {}'.format(val_y.shape))
dtrain = xgboost.DMatrix(X, label=y)
dval = xgboost.DMatrix(val_X, label=val_y)
hook = Hook.create_from_json_file()
hook.train_data = dtrain
hook.validation_data = dval
watchlist = [(dtrain, "train"), (dval, "validation")]
params = {
"max_depth": args.max_depth,
"eta": args.eta,
"gamma": args.gamma,
"min_child_weight": args.min_child_weight,
"silent": args.silent,
"objective": args.objective,
"eval_metric": args.eval_metric
}
bst = xgboost.train(params=params,
dtrain=dtrain,
evals=watchlist,
num_boost_round=args.num_round,
callbacks=[hook])
model_dir = os.environ.get('SM_MODEL_DIR')
pkl.dump(bst, open(model_dir + '/model.bin', 'wb'))
def helper_xgboost_tests(collection, save_config):
coll_name, coll_regex = collection
run_id = "trial_" + coll_name + "-" + datetime.now().strftime("%Y%m%d-%H%M%S%f")
trial_dir = os.path.join(SMDEBUG_XG_HOOK_TESTS_DIR, run_id)
hook = XG_Hook(
out_dir=trial_dir,
include_collections=[coll_name],
save_config=save_config,
export_tensorboard=True,
)
simple_xg_model(hook)
hook.close()
saved_scalars = ["scalar/xg_num_steps", "scalar/xg_before_train", "scalar/xg_after_train"]
verify_files(trial_dir, save_config, saved_scalars)
def helper_xgboost_tests(collection, save_config, with_timestamp):
coll_name, coll_regex = collection
run_id = "trial_" + coll_name + "-" + datetime.now().strftime(
"%Y%m%d-%H%M%S%f")
trial_dir = os.path.join(SMDEBUG_XG_HOOK_TESTS_DIR, run_id)
hook = XG_Hook(
out_dir=trial_dir,
include_collections=[coll_name],
save_config=save_config,
export_tensorboard=True,
)
saved_scalars = simple_xg_model(hook, with_timestamp=with_timestamp)
hook.close()
verify_files(trial_dir, save_config, saved_scalars)
if with_timestamp:
check_tf_events(trial_dir, saved_scalars)
def test_hook_save_config_collections(tmpdir):
out_dir = os.path.join(tmpdir, str(uuid.uuid4()))
hook = Hook(out_dir=out_dir,
include_collections=["metrics", "feature_importance"])
hook.get_collection("metrics").save_config = SaveConfig(save_interval=2)
hook.get_collection("feature_importance").save_config = SaveConfig(
save_interval=3)
run_xgboost_model(hook=hook)
trial = create_trial(out_dir)
metric_steps = trial.tensor("train-rmse").steps()
assert all(step % 2 == 0 for step in metric_steps[:-1])
fimps = [
t for t in trial.tensor_names() if t.startswith("feature_importance/")
]
fimp_steps = trial.tensor(fimps[0]).steps()
assert all(step % 3 == 0 for step in fimp_steps[:-1])
def create_smdebug_hook(out_dir, train_data=None, validation_data=None, frequency=1, collections=None,):
save_config = SaveConfig(save_interval=frequency)
hook = Hook(
out_dir=out_dir,
train_data=train_data,
validation_data=validation_data,
save_config=save_config,
include_collections=collections,
)
return hook
def main():
args = parse_args()
train_files_path, validation_files_path = args.train, args.validation
train_files_list = glob.glob(train_files_path + '/*.*')
print(train_files_list)
val_files_list = glob.glob(validation_files_path + '/*.*')
print(val_files_list)
print('Loading training data...')
df_train = pd.concat(map(pd.read_csv, train_files_list))
print('Loading validation data...')
df_val = pd.concat(map(pd.read_csv, val_files_list))
print('Data loading completed.')
y = df_train.Target.values
X = df_train.drop(['Target'], axis=1).values
val_y = df_val.Target.values
val_X = df_val.drop(['Target'], axis=1).values
dtrain = xgboost.DMatrix(X, label=y)
dval = xgboost.DMatrix(val_X, label=val_y)
params = {
"max_depth": args.max_depth,
"eta": args.eta,
"gamma": args.gamma,
"min_child_weight": args.min_child_weight,
"silent": args.silent,
"objective": args.objective,
"num_class": args.num_class
}
hook = Hook.create_from_json_file()
hook.train_data = dtrain
hook.validation_data = dval
watchlist = [(dtrain, "train"), (dval, "validation")]
bst = xgboost.train(params=params,
dtrain=dtrain,
evals=watchlist,
num_boost_round=args.num_round,
callbacks=[hook])
model_dir = os.environ.get('SM_MODEL_DIR')
pkl.dump(bst, open(model_dir + '/model.bin', 'wb'))
def test_hook_tree_model(tmpdir):
np.random.seed(42)
train_data = np.random.rand(5, 10)
train_label = np.random.randint(2, size=5)
dtrain = xgboost.DMatrix(train_data, label=train_label)
params = {"objective": "binary:logistic"}
bst = xgboost.train(params, dtrain, num_boost_round=0)
df = bst.trees_to_dataframe()
out_dir = os.path.join(tmpdir, str(uuid.uuid4()))
hook = Hook(out_dir=out_dir, include_collections=["trees"])
run_xgboost_model(hook=hook)
trial = create_trial(out_dir)
tensors = trial.tensor_names()
assert len(tensors) > 0
assert "trees" in trial.collections()
for col in df.columns:
assert "trees/{}".format(col) in tensors
def test_hook_save_all(tmpdir):
save_config = SaveConfig(save_steps=[0, 1, 2, 3])
out_dir = os.path.join(tmpdir, str(uuid.uuid4()))
hook = Hook(out_dir=out_dir, save_config=save_config, save_all=True)
run_xgboost_model(hook=hook)
trial = create_trial(out_dir)
collections = trial.collections()
tensors = trial.tensor_names()
assert len(tensors) > 0
assert len(trial.steps()) == 4
assert "all" in collections
assert "metrics" in collections
assert "feature_importance" in collections
assert "train-rmse" in tensors
assert any(t.startswith("feature_importance/") for t in tensors)
assert any(t.startswith("trees/") for t in tensors)
assert len(collections["all"].tensor_names) == len(tensors)
def test_hook_feature_importance(tmpdir):
np.random.seed(42)
train_data = np.random.rand(10, 10)
train_label = np.random.randint(2, size=10)
dtrain = xgboost.DMatrix(train_data, label=train_label)
out_dir = os.path.join(tmpdir, str(uuid.uuid4()))
hook = Hook(out_dir=out_dir, include_collections=["feature_importance"])
run_xgboost_model(hook=hook)
trial = create_trial(out_dir)
tensors = trial.tensor_names()
assert len(tensors) > 0
assert "feature_importance" in trial.collections()
assert any(t.startswith("feature_importance/") for t in tensors)
assert any(t.startswith("feature_importance/weight/") for t in tensors)
assert any(t.startswith("feature_importance/gain/") for t in tensors)
assert any(t.startswith("feature_importance/cover/") for t in tensors)
assert any(t.startswith("feature_importance/total_gain/") for t in tensors)
assert any(
t.startswith("feature_importance/total_cover/") for t in tensors)
def add_debugging(callbacks, hyperparameters, train_dmatrix,
val_dmatrix=None, json_config_path=None):
"""Add a sagemaker debug hook to a list of callbacks.
:param callbacks: List of callback functions.
:param hyperparameters: Dict of hyperparamters.
Same as `params` in xgb.train(params, dtrain).
:param train_dmatrix: Training data set.
:param val_dmatrix: Validation data set.
:param json_config_path: If specified, this json config will be used
instead of default config file.
"""
try:
hook = Hook.hook_from_config(json_config_path)
hook.hyperparameters = hyperparameters
hook.train_data = train_dmatrix
if val_dmatrix is not None:
hook.validation_data = val_dmatrix
callbacks.append(hook)
logging.info("Debug hook created from config")
except Exception as e:
logging.debug("Failed to create debug hook", e)
return
def test_hook_params(tmpdir):
np.random.seed(42)
train_data = np.random.rand(5, 10)
train_label = np.random.randint(2, size=5)
dtrain = xgboost.DMatrix(train_data, label=train_label)
valid_data = np.random.rand(5, 10)
valid_label = np.random.randint(2, size=5)
dvalid = xgboost.DMatrix(valid_data, label=valid_label)
params = {"objective": "binary:logistic", "num_round": 20, "eta": 0.1}
out_dir = os.path.join(tmpdir, str(uuid.uuid4()))
hook = Hook(out_dir=out_dir,
include_collections=["hyperparameters"],
hyperparameters=params)
run_xgboost_model(hook=hook)
trial = create_trial(out_dir)
tensors = trial.tensor_names()
assert len(tensors) > 0
assert "hyperparameters" in trial.collections()
assert trial.tensor("hyperparameters/objective").value(
0) == "binary:logistic"
assert trial.tensor("hyperparameters/num_round").value(0) == 20
assert trial.tensor("hyperparameters/eta").value(0) == 0.1
def test_setting_mode(tmpdir):
out_dir = os.path.join(tmpdir, str(uuid.uuid4()))
hook = Hook(out_dir=out_dir, export_tensorboard=True)
hook.set_mode(modes.GLOBAL)
with pytest.raises(ValueError):
hook.set_mode("a")
def test_hook_tensorboard_dir_created(tmpdir):
out_dir = os.path.join(tmpdir, str(uuid.uuid4()))
hook = Hook(out_dir=out_dir, export_tensorboard=True)
run_xgboost_model(hook=hook)
assert "tensorboard" in os.listdir(out_dir)
def test_hook(tmpdir):
save_config = SaveConfig(save_steps=[0, 1, 2, 3])
out_dir = os.path.join(tmpdir, str(uuid.uuid4()))
hook = Hook(out_dir=out_dir, save_config=save_config)
assert has_training_ended(out_dir) is False
run_xgboost_model(hook=hook)
