def _cuda_fit(self, X, Y, bfs_threshold, remain_trees, lock):
self._cuda_forest = cdRF(n_estimators = 1,
bootstrap = self.bootstrap,
max_features = self.max_features)
#allocate resource
self._cuda_forest.fit_init(X, Y)
f = self._cuda_forest
while True:
lock.acquire()
if remain_trees.value == 0:
lock.release()
break
remain_trees.value -= 1
lock.release()
tree = RandomClassifierTree(f)
si, n_samples = f._get_sorted_indices(f.sorted_indices)
tree.fit(f.samples, f.target, si, n_samples)
f._trees.append(tree)
#release the resource
self._cuda_forest.fit_release()
def _cuda_fit(self, X, Y, bfs_threshold, remain_trees, lock):
self._cuda_forest = cdRF(n_estimators = 1,
bootstrap = self.bootstrap,
max_features = self.max_features)
#allocate resource
self._cuda_forest.fit_init(X, Y)
f = self._cuda_forest
while True:
lock.acquire()
if remain_trees.value == 0:
lock.release()
break
remain_trees.value -= 1
lock.release()
tree = RandomClassifierTree(f)
si, n_samples = f._get_sorted_indices(f.sorted_indices)
tree.fit(f.samples, f.target, si, n_samples)
f._trees.append(tree)
#release the resource
self._cuda_forest.fit_release()
def run(self):
reload(util)
driver.init()
ctx = driver.Device(self.gpu_id).make_context()
X = self.X
Y = self.Y
bootstrap = self.bootstrap
max_features = self.max_features
bfs_threshold = self.bfs_threshold
remain_trees = self.remain_trees
lock = self.lock
trees = list()
forest = cdRF(n_estimators=1,
bootstrap=bootstrap,
max_features=max_features)
forest.fit_init(X, Y)
while True:
lock.acquire()
if remain_trees.value == 0:
lock.release()
break
remain_trees.value -= 1
lock.release()
util.log_info("CUDA gets 1 job")
tree = RandomClassifierTree(forest)
si, n_samples = forest._get_sorted_indices(forest.sorted_indices)
tree.fit(forest.samples, forest.target, si, n_samples)
trees.append(tree)
forest.fit_release()
self.result_queue.put(trees)
ctx.detach()
del ctx
util.log_info("GPU DONE")
def run(self):
reload(util)
driver.init()
ctx = driver.Device(self.gpu_id).make_context()
X = self.X
Y = self.Y
bootstrap = self.bootstrap
max_features = self.max_features
bfs_threshold = self.bfs_threshold
remain_trees = self.remain_trees
lock = self.lock
trees = list()
forest = cdRF(n_estimators = 1,
bootstrap = bootstrap,
max_features = max_features)
forest.fit_init(X, Y)
while True:
lock.acquire()
if remain_trees.value == 0:
lock.release()
break
remain_trees.value -= 1
lock.release()
util.log_info("CUDA gets 1 job")
tree = RandomClassifierTree(forest)
si, n_samples = forest._get_sorted_indices(forest.sorted_indices)
tree.fit(forest.samples, forest.target, si, n_samples)
trees.append(tree)
forest.fit_release()
self.result_queue.put(trees)
ctx.detach()
del ctx
util.log_info("GPU DONE")
