gaspersic/triton
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Tuner: made the auto-tuner compatible with the new python API

Browse Source
This commit is contained in:
Philippe Tillet
2015-08-12 19:11:10 -07:00
parent cd51f75eb5
commit b397d5306e
4 changed files with 14 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
tune/model.py
View File

@@ -30,7 +30,7 @@ def train(X, Y, profiles):
X = X[p,:]
Y = Y[p,:]
#Train the model
#Train the.profile
cut = int(1.00*M)
CV = .1
XTr, YTr = X[:,:], Y[:,:]

8
tune/optimize.py
View File

@@ -14,10 +14,10 @@ from numpy import cumsum
import tools
fetch_types = [isc.templates.fetching_policy_type.FETCH_FROM_GLOBAL_CONTIGUOUS,
isc.templates.fetching_policy_type.FETCH_FROM_GLOBAL_STRIDED,
isc.templates.fetching_policy_type.FETCH_FROM_LOCAL,
isc.templates.fetching_policy_type.FETCH_FROM_LOCAL]
fetch_types = [isc.templates.FETCH_FROM_GLOBAL_CONTIGUOUS,
isc.templates.FETCH_FROM_GLOBAL_STRIDED,
isc.templates.FETCH_FROM_LOCAL,
isc.templates.FETCH_FROM_LOCAL]
def exhaustive(template, sizes, context):
tree, _ = tools.tree_of(template, sizes, context)

4
tune/tools.py
View File

@@ -21,13 +21,13 @@ def expspace(a,b,N,r=128):
def benchmark(template, setting, tree):
queue = tree.context.queues[0]
queue.models[template, isc.float32] = isc.model(isc.float32, template(*setting), queue)
queue.profiles[template, isc.float32] = isc.profile(template(*setting), isc.float32, queue)
times = []
total = 0
i = 0
while total < 1e-2:
#z = isc.zeros(1, 10000000, isc.float32, tree.context)
z, events = isc.enqueue(tree)
z, events = isc.driver.enqueue(tree)
tree.context.queues[0].synchronize()
times.append(1e-9*sum([e.elapsed_time for e in events]))
total += times[-1]

14
tune/tune.py
View File

@@ -22,8 +22,8 @@ def pow2range(a, b):
def tune(device, operation, json_path):
#List devices
platforms = isc.get_platforms()
context = isc.context(device)
platforms = isc.driver.get_platforms()
context = isc.driver.context(device)
#List of size tuples to use
sizes = {}
@@ -83,7 +83,7 @@ def tune(device, operation, json_path):
predicted = profiles[0]
else:
clf = ensemble.RandomForestRegressor(min(10, idx+1), max_depth=min(10, idx+1)).fit(X, Y)
#clf, nrmse = model.train(X, Y, profiles)
#clf, nrmse = profile.train(X, Y, profiles)
predperf = clf.predict(x)[0]
best = (-predperf).argsort()[:5]
perf = [performance(x, tools.benchmark(operation, profiles[b], tree)) for b in best]
@@ -130,7 +130,7 @@ def tune(device, operation, json_path):
json_data[operation_name]['float32'] = {}
D = json_data[operation_name]['float32']
if len(profiles) > 1:
clf, nrmse = model.train(X, Y, profiles)
clf, nrmse = profile.train(X, Y, profiles)
D['predictor'] = [{'children_left': e.tree_.children_left.tolist(),
'children_right': e.tree_.children_right.tolist(),
'threshold': e.tree_.threshold.astype('float64').tolist(),
@@ -141,7 +141,7 @@ def tune(device, operation, json_path):
def parse_arguments():
platforms = isc.get_platforms()
platforms = isc.driver.get_platforms()
devices = [d for platform in platforms for d in platform.get_devices()]
#Command line arguments
parser = argparse.ArgumentParser()
@@ -156,7 +156,7 @@ def parse_arguments():
print("----------------")
for (i, d) in enumerate(devices):
selected = '[' + ('x' if device==d else ' ') + ']'
print selected , '-', isc.device_type_to_string(d.type), '-', d.name, 'on', d.platform.name
print selected , '-', isc.driver.device_type_to_string(d.type), '-', d.name, 'on', d.platform.name
print("----------------")
@@ -169,7 +169,7 @@ def parse_arguments():
if __name__ == "__main__":
isc.state.queue_properties = isc.CL_QUEUE_PROFILING_ENABLE
isc.driver.default.queue_properties = isc.driver.PROFILING_ENABLE
args = parse_arguments()
tune(*args)