Rajiv Shah
commited on
Commit
•
e87dc49
1
Parent(s):
ceefbb2
add py script
Browse files- H2O_Example.py +47 -0
H2O_Example.py
ADDED
@@ -0,0 +1,47 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# %%
|
2 |
+
import h2o
|
3 |
+
|
4 |
+
# %%
|
5 |
+
h2o.__version__
|
6 |
+
|
7 |
+
# %%
|
8 |
+
h2o.init()
|
9 |
+
|
10 |
+
# %%
|
11 |
+
from h2o.estimators import H2OGradientBoostingEstimator
|
12 |
+
h2o.init(jvm_custom_args=["sys.ai.h2o.debug.allowJavaVersions", "18"])
|
13 |
+
|
14 |
+
# Import the prostate dataset into H2O:
|
15 |
+
prostate = h2o.import_file("http://s3.amazonaws.com/h2o-public-test-data/smalldata/prostate/prostate.csv")
|
16 |
+
|
17 |
+
# Set the predictors and response; set the factors:
|
18 |
+
prostate["CAPSULE"] = prostate["CAPSULE"].asfactor()
|
19 |
+
predictors = ["ID","AGE","RACE","DPROS","DCAPS","PSA","VOL","GLEASON"]
|
20 |
+
response = "CAPSULE"
|
21 |
+
|
22 |
+
# Build and train the model:
|
23 |
+
pros_gbm = H2OGradientBoostingEstimator(nfolds=5,
|
24 |
+
seed=1111,
|
25 |
+
keep_cross_validation_predictions = True)
|
26 |
+
pros_gbm.train(x=predictors, y=response, training_frame=prostate)
|
27 |
+
|
28 |
+
# Eval performance:
|
29 |
+
perf = pros_gbm.model_performance()
|
30 |
+
|
31 |
+
# Generate predictions on a test set (if necessary):
|
32 |
+
pred = pros_gbm.predict(prostate)
|
33 |
+
|
34 |
+
# Extract feature interactions:
|
35 |
+
feature_interactions = pros_gbm.feature_interaction()
|
36 |
+
|
37 |
+
# %%
|
38 |
+
feature_interactions
|
39 |
+
|
40 |
+
# %%
|
41 |
+
#save model
|
42 |
+
h2o.save_model(model=pros_gbm, force=True)
|
43 |
+
|
44 |
+
# %%
|
45 |
+
pros_gbm.save_mojo('mojo')
|
46 |
+
|
47 |
+
|