def __init__(self,
booster: Booster,
tree_index: int,
x_data,
y_data,
feature_names: List[str] = None,
target_name: str = None,
class_names: (List[str], Mapping[int, str]) = None):
if hasattr(booster, 'get_booster'):
booster = booster.get_booster(
) # support XGBClassifier and XGBRegressor
utils.check_tree_index(tree_index, len(booster.get_dump()))
self.booster = booster
self.tree_index = tree_index
self.tree_to_dataframe = self._get_tree_dataframe()
self.children_left = self._calculate_children(
self.__class__.LEFT_CHILDREN_COLUMN)
self.children_right = self._calculate_children(
self.__class__.RIGHT_CHILDREN_COLUMN)
self.config = json.loads(self.booster.save_config())
self.node_to_samples = None # lazy initialized
self.features = None # lazy initialized
super().__init__(booster, x_data, y_data, feature_names, target_name,
class_names)
def load_checkpoint(checkpoint_dir, max_try=5):
"""
:param checkpoint_dir: e.g., /opt/ml/checkpoints
:param max_try: number of times to try loading checkpoint before giving up.
:return xgb_model: file path of stored xgb model. None if no checkpoint.
:return iteration: iterations completed before last checkpoiint.
"""
if not checkpoint_dir or not os.path.exists(checkpoint_dir):
return None, 0
regex = r"^{0}\.[0-9]+$".format(CHECKPOINT_FILENAME)
checkpoints = [f for f in os.listdir(checkpoint_dir) if re.match(regex, f)]
if not checkpoints:
return None, 0
checkpoints.sort()
xgb_model, iteration = None, 0
for _ in range(max_try):
try:
latest_checkpoint = checkpoints.pop()
xgb_model = os.path.join(checkpoint_dir, latest_checkpoint)
booster = Booster()
booster.load_model(xgb_model)
filename, extension = latest_checkpoint.split('.')
iteration = int(extension) + 1
break
except XGBoostError:
logging.debug("Wrong checkpoint model format %s",
latest_checkpoint)
return xgb_model, iteration
class BreastCancerTrainable(Trainable):
def setup(self, config):
self.config = config
self.nthread = config.pop("nthread", 1)
self.model: xgb.Booster = None
# Load dataset
data, labels = sklearn.datasets.load_breast_cancer(return_X_y=True)
# Split into train and test set
train_x, test_x, train_y, test_y = train_test_split(
data, labels, test_size=0.25
)
# Build input matrices for XGBoost
self.train_set = xgb.DMatrix(train_x, label=train_y)
self.test_set = xgb.DMatrix(test_x, label=test_y)
def step(self):
# you can also obtain current trial resources:
current_resources = self.trial_resources
if isinstance(current_resources, PlacementGroupFactory):
self.nthread = current_resources.head_cpus
else:
self.nthread = current_resources.cpu
results = {}
config = self.config.copy()
config["nthread"] = int(self.nthread)
self.model = xgb.train(
config,
self.train_set,
evals=[(self.test_set, "eval")],
verbose_eval=False,
xgb_model=self.model,
evals_result=results,
num_boost_round=1,
)
print(config, results)
return {"eval-logloss": results["eval"]["logloss"][-1], "nthread": self.nthread}
def save_checkpoint(self, checkpoint_dir):
path = os.path.join(checkpoint_dir, "checkpoint")
with open(path, "wb") as outputFile:
pickle.dump((self.config, self.nthread, self.model.save_raw()), outputFile)
return path
def load_checkpoint(self, checkpoint_path):
with open(checkpoint_path, "rb") as inputFile:
self.config, self.nthread, raw_model = pickle.load(inputFile)
self.model = Booster()
self.model.load_model(bytearray(raw_model))
data, labels = sklearn.datasets.load_breast_cancer(return_X_y=True)
# Split into train and test set
train_x, test_x, train_y, test_y = train_test_split(
data, labels, test_size=0.25
)
# Build input matrices for XGBoost
self.train_set = xgb.DMatrix(train_x, label=train_y)
self.test_set = xgb.DMatrix(test_x, label=test_y)
def deserialize_booster(ser_model_string):
"""
Deserialize an xgboost.core.Booster from the input ser_model_string.
"""
booster = Booster()
# TODO: change to use string io
tmp_file_name = os.path.join(_get_or_create_tmp_dir(),
f"{uuid.uuid4()}.json")
with open(tmp_file_name, "w", encoding="utf-8") as f:
f.write(ser_model_string)
booster.load_model(tmp_file_name)
return booster
def load_checkpoint(self, checkpoint_path):
with open(checkpoint_path, "rb") as inputFile:
self.config, self.nthread, raw_model = pickle.load(inputFile)
self.model = Booster()
self.model.load_model(bytearray(raw_model))
data, labels = sklearn.datasets.load_breast_cancer(return_X_y=True)
# Split into train and test set
train_x, test_x, train_y, test_y = train_test_split(data,
labels,
test_size=0.25)
# Build input matrices for XGBoost
self.train_set = xgb.DMatrix(train_x, label=train_y)
self.test_set = xgb.DMatrix(test_x, label=test_y)
class CVPack(object):
""""Auxiliary datastruct to hold one fold of CV."""
def __init__(self, dtrain, dtest, param):
""""Initialize the CVPack"""
self.dtrain = dtrain
self.dtest = dtest
self.watchlist = [(dtrain, 'train'), (dtest, 'test')]
self.bst = Booster(param, [dtrain, dtest])
def update(self, iteration, fobj):
""""Update the boosters for one iteration"""
self.bst.update(self.dtrain, iteration, fobj)
def eval(self, iteration, feval):
""""Evaluate the CVPack for one iteration."""
return self.bst.eval_set(self.watchlist, iteration, feval)
def __setstate__(self, state):
# backward compatiblity code
# load booster from raw if it is raw
# the booster now support pickle
bst = state["_Booster"]
if bst is not None and not isinstance(bst, Booster):
state["_Booster"] = Booster(model_file=bst)
self.__dict__.update(state)
def worker(model, h, v, output_path, proc_que, out_q):
params = {'nthread': 1}
model = Booster(params=params, model_file=model)
while True:
chip_x, chip_y = proc_que.get()
if chip_x == 'kill':
log.debug('Received kill')
out_q.put('killed')
break
outfile = os.path.join(
output_path,
'H{:02d}V{:02d}_{}_{}_class.p'.format(h, v, chip_x, chip_y))
if os.path.exists(outfile):
log.debug('Output exists, skipping')
try:
log.debug('Getting JSON data')
jn = get_jsonchip(h, v, chip_x, chip_y).flatten()
log.debug('Getting Aux data')
aux = get_aux(chip_x,
chip_y,
exclude=['trends', 'nlcd2001', 'nlcd2011'])
log.debug('Classifying')
results = []
for i in range(jn.shape[0]):
ccd_models = unpack_result(jn[i])
dat = np.hstack((
aux['dem'][i],
aux['aspect'][i],
aux['slope'][i],
aux['posidex'][i],
aux['mpw'][i],
))
res = []
for ccd_model in ccd_models:
res.extend(classifyccd(ccd_model, model, dat))
results.append(res)
log.debug('Saving file {}'.format(outfile))
pickle.dump(results, open(outfile, 'wb'))
except Exception as e:
log.debug('EXCEPTION with {} {}'.format(chip_x, chip_y))
log.exception(e)
def _create_checkpoint(model: Booster, epoch: int, filename: str, frequency: int):
if epoch % frequency > 0 or (not epoch and frequency > 1):
# Skip 0th checkpoint if frequency > 1
return
with tune.checkpoint_dir(step=epoch) as checkpoint_dir:
model.save_model(os.path.join(checkpoint_dir, filename))
def _train_internal(params,
dtrain,
num_boost_round=10,
evals=(),
obj=None,
feval=None,
xgb_model=None,
callbacks=None):
"""internal training function"""
callbacks = [] if callbacks is None else callbacks
evals = list(evals)
if isinstance(params, dict) \
and 'eval_metric' in params \
and isinstance(params['eval_metric'], list):
params = dict((k, v) for k, v in params.items())
eval_metrics = params['eval_metric']
params.pop("eval_metric", None)
params = list(params.items())
for eval_metric in eval_metrics:
params += [('eval_metric', eval_metric)]
bst = Booster(params, [dtrain] + [d[0] for d in evals])
nboost = 0
num_parallel_tree = 1
if xgb_model is not None:
if not isinstance(xgb_model, STRING_TYPES):
xgb_model = xgb_model.save_raw()
bst = Booster(params, [dtrain] + [d[0] for d in evals],
model_file=xgb_model)
nboost = len(bst.get_dump())
_params = dict(params) if isinstance(params, list) else params
if 'num_parallel_tree' in _params:
num_parallel_tree = _params['num_parallel_tree']
nboost //= num_parallel_tree
if 'num_class' in _params:
nboost //= _params['num_class']
# Distributed code: Load the checkpoint from rabit.
version = bst.load_rabit_checkpoint()
assert (rabit.get_world_size() != 1 or version == 0)
rank = rabit.get_rank()
start_iteration = int(version / 2)
nboost += start_iteration
callbacks_before_iter = [
cb for cb in callbacks if cb.__dict__.get('before_iteration', False)
]
callbacks_after_iter = [
cb for cb in callbacks
if not cb.__dict__.get('before_iteration', False)
]
for i in range(nboost, num_boost_round):
for cb in callbacks_before_iter:
cb(
CallbackEnv(model=bst,
cvfolds=None,
iteration=i,
begin_iteration=start_iteration,
end_iteration=num_boost_round,
rank=rank,
evaluation_result_list=None))
# Distributed code: need to resume to this point.
# Skip the first update if it is a recovery step.
if version % 2 == 0:
bst.update(dtrain, i, obj)
bst.save_rabit_checkpoint()
version += 1
assert (rabit.get_world_size() == 1
or version == rabit.version_number())
nboost += 1
evaluation_result_list = []
# check evaluation result.
if len(evals) != 0:
bst_eval_set = bst.eval_set(evals, i, feval)
if isinstance(bst_eval_set, STRING_TYPES):
msg = bst_eval_set
else:
msg = bst_eval_set.decode()
res = [x.split(':') for x in msg.split()]
evaluation_result_list = [(k, float(v)) for k, v in res[1:]]
try:
for cb in callbacks_after_iter:
cb(
CallbackEnv(model=bst,
cvfolds=None,
iteration=i,
begin_iteration=start_iteration,
end_iteration=num_boost_round,
rank=rank,
evaluation_result_list=evaluation_result_list))
except EarlyStopException:
break
# do checkpoint after evaluation, in case evaluation also updates booster.
bst.save_rabit_checkpoint()
version += 1
if bst.attr('best_score') is not None:
bst.best_score = float(bst.attr('best_score'))
bst.best_iteration = int(bst.attr('best_iteration'))
else:
bst.best_iteration = nboost - 1
bst.best_ntree_limit = (bst.best_iteration + 1) * num_parallel_tree
return bst
def __init__(self, dtrain, dtest, param):
""""Initialize the CVPack"""
self.dtrain = dtrain
self.dtest = dtest
self.watchlist = [(dtrain, 'train'), (dtest, 'test')]
self.bst = Booster(param, [dtrain, dtest])
def _train_internal(params, dtrain,
num_boost_round=10, evals=(),
obj=None, feval=None,
xgb_model=None, callbacks=None):
"""internal training function"""
callbacks = [] if callbacks is None else callbacks
evals = list(evals)
if isinstance(params, dict) \
and 'eval_metric' in params \
and isinstance(params['eval_metric'], list):
params = dict((k, v) for k, v in params.items())
eval_metrics = params['eval_metric']
params.pop("eval_metric", None)
params = list(params.items())
for eval_metric in eval_metrics:
params += [('eval_metric', eval_metric)]
bst = Booster(params, [dtrain] + [d[0] for d in evals])
nboost = 0
num_parallel_tree = 1
if xgb_model is not None:
if not isinstance(xgb_model, STRING_TYPES):
xgb_model = xgb_model.save_raw()
bst = Booster(params, [dtrain] + [d[0] for d in evals],
model_file=xgb_model)
nboost = len(bst.get_dump())
_params = dict(params) if isinstance(params, list) else params
if 'num_parallel_tree' in _params:
num_parallel_tree = _params['num_parallel_tree']
nboost //= num_parallel_tree
if 'num_class' in _params:
nboost //= _params['num_class']
# Distributed code: Load the checkpoint from rabit.
version = bst.load_rabit_checkpoint()
assert (rabit.get_world_size() != 1 or version == 0)
rank = rabit.get_rank()
start_iteration = int(version / 2)
nboost += start_iteration
callbacks_before_iter = [
cb for cb in callbacks if cb.__dict__.get('before_iteration', False)]
callbacks_after_iter = [
cb for cb in callbacks if
not cb.__dict__.get('before_iteration', False)]
for i in range(nboost, num_boost_round):
for cb in callbacks_before_iter:
cb(CallbackEnv(model=bst,
cvfolds=None,
iteration=i,
begin_iteration=start_iteration,
end_iteration=num_boost_round,
rank=rank,
evaluation_result_list=None))
# Distributed code: need to resume to this point.
# Skip the first update if it is a recovery step.
if version % 2 == 0:
bst.update(dtrain, i, obj)
bst.save_rabit_checkpoint()
version += 1
assert (
rabit.get_world_size() == 1 or version == rabit.version_number())
nboost += 1
evaluation_result_list = []
# check evaluation result.
if len(evals) != 0:
bst_eval_set = bst.eval_set(evals, i, feval)
if isinstance(bst_eval_set, STRING_TYPES):
msg = bst_eval_set
else:
msg = bst_eval_set.decode()
res = [x.split(':') for x in msg.split()]
evaluation_result_list = [(k, float(v)) for k, v in res[1:]]
try:
for cb in callbacks_after_iter:
cb(CallbackEnv(model=bst,
cvfolds=None,
iteration=i,
begin_iteration=start_iteration,
end_iteration=num_boost_round,
rank=rank,
evaluation_result_list=evaluation_result_list))
except EarlyStopException:
break
# do checkpoint after evaluation, in case evaluation also updates booster.
bst.save_rabit_checkpoint()
version += 1
if bst.attr('best_score') is not None:
bst.best_score = float(bst.attr('best_score'))
bst.best_iteration = int(bst.attr('best_iteration'))
else:
bst.best_iteration = nboost - 1
bst.best_ntree_limit = (bst.best_iteration + 1) * num_parallel_tree
return bst
if _pandas:
from xgboost.core import Booster, DMatrix
else:
sys.modules['pandas'] = None
from xgboost.core import Booster, DMatrix
del sys.modules['pandas']
DEBUG = False
MODEL_NAME = 'v12.xgb'
PredictionRow = namedtuple('PredictionRow', ('c', 't', 's', 'p'))
handles = partial(register_handler, 'editor')
doc_generator = DocGenerator()
with resources.path('akimous.resources', MODEL_NAME) as _path:
model = Booster(model_file=str(_path)) # 3 ms
model.set_param('nthread', 1)
logger.info('Model %s loaded.', MODEL_NAME)
def get_relative_path(context):
try:
return tuple(
context.path.relative_to(context.shared.project_root).parts)
except ValueError:
# the file does not belong to the project folder
return tuple(context.path.parts)
async def run_pylint(context, send):
if not config['linter']['pylint']: