def onCvresults(self, *args, **kwargs):
del kwargs["how"]
self._writePreamble("CV Test Results")
text = "Display the results of a cross-validation test with the " \
"current settings against the defaults."
nsets = options["TestToolsUI", "n"]
# With defaults first.
self.write("<p>Testing with defaults...</p>")
saved = {}
for opt in options.options(True):
# Ignore those that have do_not_restore as True
# (These are predominately storage options, and at least
# the cache directory ones may be needed later on).
sect, opt = opt[1:].split(']', 1)
saved[(sect, opt)] = options[(sect, opt)]
if not options.no_restore(sect, opt):
options.set(sect, opt, options.default(sect, opt))
options["TestToolsUI", "source"] = kwargs["TestToolsUI_source"]
# XXX Cache this somewhere? If the testing data isn't changing,
# XXX and the user is running multiple tests, then it doesn't
# XXX make much sense to rerun the 'default's test over and over
# XXX again.
cv_out, errors = self.timCV(nsets)
## print errors.read()
defaults = self.rates(cv_out)
# Now with specified settings.
self.write("<p>Testing with selected settings...</p>")
for opt in options.options(True):
sect, opt = opt[1:].split(']', 1)
try:
value = kwargs["%s_%s" % (sect, opt)]
except KeyError:
# Leave as the default.
pass
else:
options.set(sect, opt, value)
cv_out, errors = self.timCV(nsets)
## print errors.read()
current = self.rates(cv_out)
# Restore the settings.
for opt in options.options(True):
sect, opt = opt[1:].split(']', 1)
options.set(sect, opt, saved[(sect, opt)])
# Do the comparison.
comp, errors = self.compare(defaults, current)
## print errors.read()
# Output the results
# XXX This is just what you'd get from running cmp.py
# XXX at the moment - it could be prettied up a bit.
comp = comp.read()
box = self._buildBox('Cross-validation test', None,
cgi.escape(comp).replace("\n", "<br />"))
self.write(box)
self._writePostamble()
def onCvresults(self, *args, **kwargs):
del kwargs["how"]
self._writePreamble("CV Test Results")
text = "Display the results of a cross-validation test with the " \
"current settings against the defaults."
nsets = options["TestToolsUI", "n"]
# With defaults first.
self.write("<p>Testing with defaults...</p>")
saved = {}
for opt in options.options(True):
# Ignore those that have do_not_restore as True
# (These are predominately storage options, and at least
# the cache directory ones may be needed later on).
sect, opt = opt[1:].split(']', 1)
saved[(sect, opt)] = options[(sect, opt)]
if not options.no_restore(sect, opt):
options.set(sect, opt, options.default(sect, opt))
options["TestToolsUI", "source"] = kwargs["TestToolsUI_source"]
# XXX Cache this somewhere? If the testing data isn't changing,
# XXX and the user is running multiple tests, then it doesn't
# XXX make much sense to rerun the 'default's test over and over
# XXX again.
cv_out, errors = self.timCV(nsets)
## print errors.read()
defaults = self.rates(cv_out)
# Now with specified settings.
self.write("<p>Testing with selected settings...</p>")
for opt in options.options(True):
sect, opt = opt[1:].split(']', 1)
try:
value = kwargs["%s_%s" % (sect, opt)]
except KeyError:
# Leave as the default.
pass
else:
options.set(sect, opt, value)
cv_out, errors = self.timCV(nsets)
## print errors.read()
current = self.rates(cv_out)
# Restore the settings.
for opt in options.options(True):
sect, opt = opt[1:].split(']', 1)
options.set(sect, opt, saved[(sect, opt)])
# Do the comparison.
comp, errors = self.compare(defaults, current)
## print errors.read()
# Output the results
# XXX This is just what you'd get from running cmp.py
# XXX at the moment - it could be prettied up a bit.
comp = comp.read()
box = self._buildBox('Cross-validation test', None,
cgi.escape(comp).replace("\n", "<br />"))
self.write(box)
self._writePostamble()
def open(db_name, mode):
if os.path.exists(db_name) and \
options.default("globals", "dbm_type") != \
options["globals", "dbm_type"]:
dbm_type = whichdb.whichdb(db_name)
if sys.platform == "win32" and sys.version_info < (2,3) and \
dbm_type == "dbhash":
dbm_type = "db3hash"
else:
dbm_type = options["globals", "dbm_type"].lower()
f = open_funcs.get(dbm_type)
if f is None:
raise error("Unknown dbm type: %s" % dbm_type)
return f(db_name, mode)
def open(db_name, mode):
if os.path.exists(db_name) and \
options.default("globals", "dbm_type") != \
options["globals", "dbm_type"]:
# let the file tell us what db to use
dbm_type = whichdb.whichdb(db_name)
# if we are using Windows and Python < 2.3, then we need to use
# db3hash, not dbhash.
if (sys.platform == "win32" and sys.version_info < (2, 3)
and dbm_type == "dbhash"):
dbm_type = "db3hash"
else:
# fresh file or overridden - open with what the user specified
dbm_type = options["globals", "dbm_type"].lower()
f = open_funcs.get(dbm_type)
if f is None:
raise error("Unknown dbm type: %s" % dbm_type)
return f(db_name, mode)
def open(db_name, mode):
if os.path.exists(db_name) and \
options.default("globals", "dbm_type") != \
options["globals", "dbm_type"]:
# let the file tell us what db to use
dbm_type = whichdb.whichdb(db_name)
# if we are using Windows and Python < 2.3, then we need to use
# db3hash, not dbhash.
if (sys.platform == "win32" and
sys.version_info < (2, 3) and
dbm_type == "dbhash"):
dbm_type = "db3hash"
else:
# fresh file or overridden - open with what the user specified
dbm_type = options["globals", "dbm_type"].lower()
f = open_funcs.get(dbm_type)
if f is None:
raise error("Unknown dbm type: %s" % dbm_type)
return f(db_name, mode)
def onCvresults(self, *args, **kwargs):
del kwargs["how"]
self._writePreamble("CV Test Results")
text = "Display the results of a cross-validation test with the " \
"current settings against the defaults."
nsets = options["TestToolsUI", "n"]
self.write("<p>Testing with defaults...</p>")
saved = {}
for opt in options.options(True):
sect, opt = opt[1:].split(']', 1)
saved[(sect, opt)] = options[(sect, opt)]
if not options.no_restore(sect, opt):
options.set(sect, opt, options.default(sect, opt))
options["TestToolsUI", "source"] = kwargs["TestToolsUI_source"]
cv_out, errors = self.timCV(nsets)
defaults = self.rates(cv_out)
self.write("<p>Testing with selected settings...</p>")
for opt in options.options(True):
sect, opt = opt[1:].split(']', 1)
try:
value = kwargs["%s_%s" % (sect, opt)]
except KeyError:
pass
else:
options.set(sect, opt, value)
cv_out, errors = self.timCV(nsets)
current = self.rates(cv_out)
for opt in options.options(True):
sect, opt = opt[1:].split(']', 1)
options.set(sect, opt, saved[(sect, opt)])
comp, errors = self.compare(defaults, current)
comp = comp.read()
box = self._buildBox('Cross-validation test', None,
cgi.escape(comp).replace("\n", "<br />"))
self.write(box)
self._writePostamble()
import getopt
from spambayes.Options import options, get_pathname_option
from spambayes import mboxutils, hammie, Corpus, storage
Corpus.Verbose = True
program = sys.argv[0] # For usage(); referenced by docstring above
# Default database name
# This is a bit of a hack to counter the default for
# persistent_storage_file changing from ~/.hammiedb to hammie.db
# This will work unless a user had hammie.db as their value for
# persistent_storage_file
if options["Storage", "persistent_storage_file"] == \
options.default("Storage", "persistent_storage_file"):
options["Storage", "persistent_storage_file"] = \
os.path.join("~", ".hammiedb")
DEFAULTDB = get_pathname_option("Storage", "persistent_storage_file")
# Probability at which a message is considered spam
SPAM_THRESHOLD = options["Categorization", "spam_cutoff"]
HAM_THRESHOLD = options["Categorization", "ham_cutoff"]
def train(h, msgs, is_spam):
"""Train bayes with all messages from a mailbox."""
mbox = mboxutils.getmbox(msgs)
i = 0
for msg in mbox:
i += 1
import getopt
from spambayes.Options import options, get_pathname_option
from spambayes import mboxutils, hammie, Corpus, storage
Corpus.Verbose = True
program = sys.argv[0] # For usage(); referenced by docstring above
# Default database name
# This is a bit of a hack to counter the default for
# persistent_storage_file changing from ~/.hammiedb to hammie.db
# This will work unless a user had hammie.db as their value for
# persistent_storage_file
if options["Storage", "persistent_storage_file"] == \
options.default("Storage", "persistent_storage_file"):
options["Storage", "persistent_storage_file"] = \
os.path.join("~", ".hammiedb")
DEFAULTDB = get_pathname_option("Storage", "persistent_storage_file")
# Probability at which a message is considered spam
SPAM_THRESHOLD = options["Categorization", "spam_cutoff"]
HAM_THRESHOLD = options["Categorization", "ham_cutoff"]
def train(h, msgs, is_spam):
"""Train bayes with all messages from a mailbox."""
mbox = mboxutils.getmbox(msgs)
i = 0
for msg in mbox:
i += 1
import getopt
from spambayes.Options import options
from spambayes import classifier, mboxutils, hammie, Corpus
Corpus.Verbose = True
program = sys.argv[0] # For usage(); referenced by docstring above
# Default database name
DEFAULTDB = os.path.expanduser(options["Storage", "persistent_storage_file"])
# This is a bit of a hack to counter the default for
# persistent_storage_file changing from ~/.hammiedb to hammie.db
# This will work unless a user had hammie.db as their value for
# persistent_storage_file
if DEFAULTDB == options.default("Storage", "persistent_storage_file"):
DEFAULTDB = os.path.expanduser(os.path.join("~", ".hammiedb"))
# Probability at which a message is considered spam
SPAM_THRESHOLD = options["Categorization", "spam_cutoff"]
HAM_THRESHOLD = options["Categorization", "ham_cutoff"]
def train(h, msgs, is_spam):
"""Train bayes with all messages from a mailbox."""
mbox = mboxutils.getmbox(msgs)
i = 0
for msg in mbox:
i += 1
if i % 10 == 0:
sys.stdout.write("\r%6d" % i)
