-
-
Save 1328/2f94044449a60f588648 to your computer and use it in GitHub Desktop.
pytables example
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import random | |
import tables | |
import time | |
import os.path | |
from pprint import pprint | |
def timer(N=1): | |
'''timing decorator''' | |
def real_deca(func): | |
def wrapper(*args, **kws): | |
ts = time.time() | |
for i in range(N): | |
res = func(*args, **kws) | |
e = time.time() - ts | |
print('{} -> Did {} runs, avg {}, total {}'.format( | |
func.__name__, | |
N, | |
e/N, | |
e | |
)) | |
return res | |
return wrapper | |
return real_deca | |
class Data(tables.IsDescription): | |
name = tables.StringCol(16) | |
num = tables.Int64Col() | |
def build(n = 10): | |
'''create random data set of length = n''' | |
ds = [] # data set | |
fs = [] # filter set | |
for i in range(n): | |
ds.append({ | |
'name':''.join(chr(random.randrange(97,120)) for _ in range(15)), | |
'num':random.randrange(1000,9999999) | |
}) | |
for i in range(n//2): | |
d = ds[i] | |
ds.append({ | |
'name':d['name'], | |
'num':random.randrange(1000,9999999), | |
}) | |
return ds | |
def build_filter(ds): | |
'''build random filter set from ds''' | |
random.shuffle(ds) | |
fs = set() | |
for i in range(len(ds)//4): | |
fs.add(ds[i]['name']) | |
random.shuffle(ds) | |
return fs | |
def write_data(dat): | |
'''write data set into pytable''' | |
fn = os.path.normpath('c:/tmp/test.h5') | |
h5file = tables.open_file(fn, mode = "w", title = "Testing") | |
group = h5file.create_group("/", 'log', 'log information') | |
table = h5file.create_table(group, 'dd', Data, "Readout example") | |
d = table.row | |
for i in dat: | |
d['name'] = i['name'] | |
d['num'] = i['num'] | |
d.append() | |
table.cols.name.create_index() | |
table.cols.num.create_index() | |
h5file.close() | |
@timer() | |
def xc_mem(ds, fs): | |
'''do compare with both data sets in memory''' | |
res = [] | |
for nm in fs: | |
candi = [d['num'] for d in ds if d['name'] == nm] | |
res.append({'name':nm, 'num':max(candi)}) | |
return res | |
@timer() | |
def xc_hd5(fs): | |
'''do compare with data set in pytables''' | |
fn = os.path.normpath('c:/tmp/test.h5') | |
h5file = tables.open_file(fn) | |
tbl = h5file.root.log.dd | |
res = [] | |
for nm in fs: | |
q = 'name == b"{}"'.format(nm) | |
candi = [d['num'] for d in tbl.read_where(q)] | |
res.append({'name':nm, 'num':max(candi)}) | |
return res | |
def main(): | |
ds = build(10000) | |
fs = build_filter(ds) | |
print('data set {}, filter set {}'.format(len(ds),len(fs))) | |
write_data(ds) | |
hr = xc_hd5(fs) | |
mr = xc_mem(ds, fs) | |
print(mr == hr) | |
main() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment