In [1]:

    

import h2o


    

In [2]:

    

# Connect to a pre-existing cluster
h2o.init()


    

    





H2O cluster uptime: 
4 minutes 56 seconds 274 milliseconds 
H2O cluster version: 
3.1.0.99999
H2O cluster name: 
ece
H2O cluster total nodes: 
1
H2O cluster total memory: 
4.44 GB
H2O cluster total cores: 
8
H2O cluster allowed cores: 
8
H2O cluster healthy: 
True
H2O Connection ip: 
127.0.0.1
H2O Connection port: 
54321

In [3]:

    

df = h2o.import_file(path=h2o.locate("smalldata/logreg/prostate.csv"))


    

    




Parse Progress: [##################################################] 100%
Imported  /Users/ece/0xdata/h2o-dev/smalldata/logreg/prostate.csv . Parsed 380 rows and 9 cols

In [4]:

    

df.describe()


    

    




Rows: 380 Cols: 9

Chunk compression summary:







    





chunk_type
chunk_name
count
count_percentage
size
size_percentage
CBS
Bits
1
11.111112
    118  B
2.4210093
C1N
1-Byte Integers (w/o NAs)
5
55.555557
    2.2 KB
45.958145
C2
2-Byte Integers
1
11.111112
    828  B
16.9881
C2S
2-Byte Fractions
2
22.222223
    1.6 KB
34.632744






    




Frame distribution summary:







    






size
number_of_rows
number_of_chunks_per_column
number_of_chunks
172.16.2.41:54321
    4.8 KB
380.0
1.0
9.0
mean
    4.8 KB
380.0
1.0
9.0
min
    4.8 KB
380.0
1.0
9.0
max
    4.8 KB
380.0
1.0
9.0
stddev
      0  B
0.0
0.0
0.0
total
    4.8 KB
380.0
1.0
9.0






    




Column-by-Column Summary:







    






ID
CAPSULE
AGE
RACE
DPROS
DCAPS
PSA
VOL
GLEASON
type
int
int
int
int
int
int
real
real
int
mins
1.0
0.0
43.0
0.0
1.0
1.0
0.3
0.0
0.0
mean
190.5
0.402631578947
66.0394736842
1.08684210526
2.27105263158
1.10789473684
15.4086315789
15.8129210526
6.38421052632
maxs
380.0
1.0
79.0
2.0
4.0
2.0
139.7
97.6
9.0
sigma
109.840793879
0.491074338963
6.52707126917
0.308773258025
1.00010761815
0.310656449351
19.9975726686
18.3476199673
1.09195337443
zero_count
0
227
0
3
0
0
0
167
2
missing_count
0
0
0
0
0
0
0
0
0

In [5]:

    

# Remove ID from training frame
train = df.drop("ID")


    

In [6]:

    

# For VOL & GLEASON, a zero really means "missing"
vol = train['VOL']
vol[vol == 0] = None
gle = train['GLEASON']
gle[gle == 0] = None


    

In [7]:

    

# Convert CAPSULE to a logical factor
train['CAPSULE'] = train['CAPSULE'].asfactor()


    

In [8]:

    

# See that the data is ready
train.describe()


    

    




Rows: 380 Cols: 8

Chunk compression summary:







    





chunk_type
chunk_name
count
count_percentage
size
size_percentage
CBS
Bits
1
12.5
    118  B
2.9164608
C1
1-Byte Integers
1
12.5
    448  B
11.072664
C1N
1-Byte Integers (w/o NAs)
4
50.0
    1.8 KB
44.290657
C2S
2-Byte Fractions
2
25.0
    1.6 KB
41.72022






    




Frame distribution summary:







    






size
number_of_rows
number_of_chunks_per_column
number_of_chunks
172.16.2.41:54321
    4.0 KB
380.0
1.0
8.0
mean
    4.0 KB
380.0
1.0
8.0
min
    4.0 KB
380.0
1.0
8.0
max
    4.0 KB
380.0
1.0
8.0
stddev
      0  B
0.0
0.0
0.0
total
    4.0 KB
380.0
1.0
8.0






    




Column-by-Column Summary:







    






CAPSULE
AGE
RACE
DPROS
DCAPS
PSA
VOL
GLEASON
type
enum
int
int
int
int
real
real
int
mins
0.0
43.0
0.0
1.0
1.0
0.3
1.92
4.0
mean
0.402631578947
66.0394736842
1.08684210526
2.27105263158
1.10789473684
15.4086315789
28.2108450704
6.41798941799
maxs
1.0
79.0
2.0
4.0
2.0
139.7
97.6
9.0
sigma
0.491074338963
6.52707126917
0.308773258025
1.00010761815
0.310656449351
19.9975726686
15.824317841
0.990616012776
zero_count
227
0
3
0
0
0
0
0
missing_count
0
0
0
0
0
0
167
2

In [9]:

    

# Run GBM
my_gbm = h2o.gbm(           y=train["CAPSULE"],
                 validation_y=train["CAPSULE"],
                            x=train[1:],
                 validation_x=train[1:],
                 distribution = "bernoulli",
                 ntrees=50,
                 learn_rate=0.1)


    

    




gbm Model Build Progress: [##################################################] 100%

In [10]:

    

my_gbm_metrics = my_gbm.model_performance(train)
my_gbm_metrics.show()


    

    




ModelMetricsBinomial: gbm
** Reported on test data. **

MSE: 0.0744435128035
R^2: 0.691303022834
LogLoss: 0.270431302856
AUC: 0.980550516829
Gini: 0.961101033659

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.452447834853:







    






0
1
Error
Rate
0
214.0
13.0
0.0573
 (13.0/227.0)
1
14.0
139.0
0.0915
 (14.0/153.0)
Total
228.0
152.0
0.1488
 (0.1488/380.0)






    




Maximum Metrics:







    





metric
threshold
value
idx
f1
0.452447834853
0.911475409836
151.0
f2
0.316330050777
0.942928039702
192.0
f0point5
0.561811969917
0.926756352765
128.0
accuracy
0.452447834853
0.928947368421
151.0
precision
0.984119707771
1.0
0.0
absolute_MCC
0.452447834853
0.852148594702
151.0
min_per_class_accuracy
0.433436616032
0.921568627451
157.0
tns
0.984119707771
227.0
0.0
fns
0.984119707771
152.0
0.0
fps
0.012063597474
227.0
378.0
tps
0.197563211765
153.0
229.0
tnr
0.984119707771
1.0
0.0
fnr
0.984119707771
0.993464052288
0.0
fpr
0.012063597474
1.0
378.0
tpr
0.197563211765
1.0
229.0