def wiki_proportions_chart(path, prefixes):
prefixes, sizes = zip(*sorted([(pr, dumpSize(pr)) for pr in prefixes],
key=operator.itemgetter(1)))
blockSize = 5
ind = p.arange(0, blockSize * len(prefixes),
blockSize) # y location for groups
height = 4 # bar height
#colors = ['g','r','c','m','y']
thresholds = [5000, 2000, 1000, 500, 200, 100, 50, 20, 10]
colors = [
str(float(i + 1) / (len(thresholds) + 1))
for i in xrange(len(thresholds))
]
colors.reverse()
p.clf()
"""
overall = p.barh( ind
, [1.0] * len(ind)
, height
, color = 'b'
, linewidth = 0
, align='center'
)
"""
subbars = []
for i, thresh in enumerate(thresholds):
subbars.append(
p.barh(ind, [
float(docs_under_thresh(pr, thresh)) / dumpSize(pr)
for pr in prefixes
],
height,
color=colors[i % len(colors)],
linewidth=0,
align='center'))
p.ylim(-height, len(prefixes) * blockSize)
p.xlim(0, 1)
yfontprop = FontProperties(size=4)
xfontprop = FontProperties(size=4)
p.xlabel('Proportion')
p.ylabel('Language Code')
p.title('Proportion of Documents Under Threshold')
p.yticks(ind, prefixes, fontproperties=yfontprop)
xmin, xmax = p.xlim()
xtick_interval = 0.1
p.xticks(p.arange(xmin, xmax, xtick_interval), fontproperties=xfontprop)
p.gca().xaxis.grid(linestyle='-', linewidth=0.15)
p.gca().yaxis.tick_left()
p.savefig(path, dpi=300)
p.close()
p.clf()
def wiki_proportions_chart(path, prefixes):
prefixes, sizes = zip(*sorted( [(pr, dumpSize(pr)) for pr in prefixes]
, key = operator.itemgetter(1)
)
)
blockSize = 5
ind = p.arange(0, blockSize*len(prefixes), blockSize) # y location for groups
height = 4 # bar height
#colors = ['g','r','c','m','y']
thresholds = [5000, 2000,1000,500,200,100,50,20,10]
colors = [str(float(i+1) / (len(thresholds)+1)) for i in xrange(len(thresholds))]
colors.reverse()
p.clf()
"""
overall = p.barh( ind
, [1.0] * len(ind)
, height
, color = 'b'
, linewidth = 0
, align='center'
)
"""
subbars = []
for i, thresh in enumerate(thresholds) :
subbars.append( p.barh( ind
, [ float(docs_under_thresh(pr, thresh)) / dumpSize(pr) for pr in prefixes]
, height
, color = colors[ i % len(colors) ]
, linewidth = 0
, align='center'
)
)
p.ylim(-height, len(prefixes) * blockSize)
p.xlim(0, 1)
yfontprop = FontProperties(size=4)
xfontprop = FontProperties(size=4)
p.xlabel('Proportion')
p.ylabel('Language Code')
p.title('Proportion of Documents Under Threshold')
p.yticks(ind, prefixes, fontproperties = yfontprop)
xmin, xmax = p.xlim()
xtick_interval = 0.1
p.xticks( p.arange(xmin,xmax,xtick_interval),fontproperties = xfontprop)
p.gca().xaxis.grid(linestyle = '-', linewidth=0.15)
p.gca().yaxis.tick_left()
p.savefig(path, dpi=300)
p.close()
p.clf()
def wiki_sizes_chart(path, prefixes, upperlimit = None ):
prefixes, sizes = zip(*sorted( [(pr, dumpSize(pr)) for pr in prefixes]
, key = operator.itemgetter(1)
)
)
blockSize = 5
ind = p.arange(0, blockSize*len(prefixes), blockSize) # y location for groups
height = 4 # bar height
#colors = ['g','r','c','m','y']
colors = html_colors
thresholds = [5000, 2000,1000,500,200,100,50,20,10]
#colors = [str(float(i+1) / (len(thresholds)+1)) for i in xrange(len(thresholds))]
#colors.reverse()
overall = p.barh( ind
, sizes
, height
, color = 'b'
, linewidth = 0
, align='center'
)
subbars = []
for i, thresh in enumerate(thresholds) :
subbars.append( p.barh( ind
, [ docs_under_thresh(pr, thresh) for pr in prefixes]
, height
, color = colors[ i % len(colors) ]
, linewidth = 0
, align='center'
)
)
p.ylim(-height, len(prefixes) * blockSize)
if upperlimit:
p.xlim(0, upperlimit)
yfontprop = FontProperties(size=4)
xfontprop = FontProperties(size=4)
p.xlabel('Documents')
p.ylabel('Language Code')
p.title('Number of Documents Under Threshold')
p.yticks(ind, prefixes, fontproperties = yfontprop)
xmin, xmax = p.xlim()
xtick_interval = rounded_interval(xmin, xmax, 20, 2)
p.xticks( p.arange(xmin,xmax,xtick_interval),fontproperties = xfontprop)
p.gca().xaxis.grid(linestyle = '-', linewidth=0.15)
p.gca().yaxis.tick_left()
p.legend( [ b[0] for b in subbars]
, map(str,thresholds)
, prop = xfontprop
, loc = 'lower right'
)
p.savefig(path, dpi=300)
p.close()
p.clf()
def wiki_sizes_chart(path, prefixes, upperlimit=None):
prefixes, sizes = zip(*sorted([(pr, dumpSize(pr)) for pr in prefixes],
key=operator.itemgetter(1)))
blockSize = 5
ind = p.arange(0, blockSize * len(prefixes),
blockSize) # y location for groups
height = 4 # bar height
#colors = ['g','r','c','m','y']
colors = html_colors
thresholds = [5000, 2000, 1000, 500, 200, 100, 50, 20, 10]
#colors = [str(float(i+1) / (len(thresholds)+1)) for i in xrange(len(thresholds))]
#colors.reverse()
overall = p.barh(ind,
sizes,
height,
color='b',
linewidth=0,
align='center')
subbars = []
for i, thresh in enumerate(thresholds):
subbars.append(
p.barh(ind, [docs_under_thresh(pr, thresh) for pr in prefixes],
height,
color=colors[i % len(colors)],
linewidth=0,
align='center'))
p.ylim(-height, len(prefixes) * blockSize)
if upperlimit:
p.xlim(0, upperlimit)
yfontprop = FontProperties(size=4)
xfontprop = FontProperties(size=4)
p.xlabel('Documents')
p.ylabel('Language Code')
p.title('Number of Documents Under Threshold')
p.yticks(ind, prefixes, fontproperties=yfontprop)
xmin, xmax = p.xlim()
xtick_interval = rounded_interval(xmin, xmax, 20, 2)
p.xticks(p.arange(xmin, xmax, xtick_interval), fontproperties=xfontprop)
p.gca().xaxis.grid(linestyle='-', linewidth=0.15)
p.gca().yaxis.tick_left()
p.legend([b[0] for b in subbars],
map(str, thresholds),
prop=xfontprop,
loc='lower right')
p.savefig(path, dpi=300)
p.close()
p.clf()
def create_dataset(num_doc,
distribution,
num_seg,
text_transformer,
exclude=[],
seed=None):
"""
Create a multilingual dataset from wikipedia data based on segments
of monolingual documents.
@param num_doc: number of documents in intended dataset
@param distribution: Mapping from class name to proportion
@param segments: Sequence of segments and their relative sizes
@param exclude: Docids to be excluded from the final dataset
"""
logger = logging.getLogger('wikidump.segment.create_dataset')
logger.info('Creating dataset')
docmap = {}
classmap = {}
# Set up a parser for each class
parser = {}
classlabels = []
class_probs = numpy.empty(len(distribution), dtype=float)
dump_indices = []
denominator = 0
logger.info('Setting up parsers')
for i, classlabel in enumerate(distribution):
parser[classlabel] = page_parser(classlabel)
classlabels.append(classlabel)
denominator += distribution[classlabel]
class_probs[i] = denominator
dumpsize = dumpSize(classlabel)
indices = range(dumpsize)
random.shuffle(indices)
dump_indices.append(indices)
logger.debug('Done for %s, %d indices', classlabel, dumpsize)
logger.info('Parsers ready')
# Normalize to a CDF
class_probs /= denominator
random.seed(seed)
logger.info('Obtaining documents')
for doc_num in range(num_doc):
doc = ""
classes = set()
segids = []
for seg_num in range(num_seg):
seg_class_index = numpy.searchsorted(class_probs, random.random())
seg_class = classlabels[seg_class_index]
classes.add(seg_class)
doc_index = dump_indices[seg_class_index].pop(0)
segid = "_".join((seg_class, str(doc_index)))
while segid in exclude:
logger.debug('Ignored %s: in exclude', segid)
try:
doc_index = dump_indices[seg_class_index].pop(0)
except IndexError:
raise ValueError, "No more documents in class %s available" % classlabel
segid = "_".join((seg_class, str(doc_index)))
segids.append(segid)
content = text_transformer(parser[seg_class].get_entry(doc_index))
seg_size = len(content) / num_seg
seg_start = seg_size * seg_num
seg_end = seg_start + seg_size
doc += content[seg_start:seg_end]
docid = '-'.join(segids)
docmap[docid] = doc
classmap[docid] = list(classes)
logger.debug('Index: %d ID: %s', doc_num, docid)
return docmap, classmap
raise ValueError, "No more documents in class %s available" % classlabel
segid = "_".join((seg_class, str(doc_index)))
segids.append(segid)
content = text_transformer(parser[seg_class].get_entry(doc_index))
seg_size = len(content) / num_seg
seg_start = seg_size * seg_num
seg_end = seg_start + seg_size
doc += content[seg_start:seg_end]
docid = '-'.join(segids)
docmap[docid] = doc
classmap[docid] = list(classes)
logger.debug('Index: %d ID: %s', doc_num, docid)
return docmap, classmap
if __name__ == "__main__":
picklefile = open('segment.pickle', "w")
distribution = {}
for p in all_prefixes:
if len(p) > 3:
logger.warning('Ignoring %s', p)
else:
distribution[p] = dumpSize(p)
data = create_dataset(5000,
distribution,
2,
strip_mediawiki_markup,
exclude=[],
seed=61383441363)
dump(data, picklefile)
def create_dataset(num_doc, distribution, num_seg, text_transformer, exclude=[], seed=None):
"""
Create a multilingual dataset from wikipedia data based on segments
of monolingual documents.
@param num_doc: number of documents in intended dataset
@param distribution: Mapping from class name to proportion
@param segments: Sequence of segments and their relative sizes
@param exclude: Docids to be excluded from the final dataset
"""
logger = logging.getLogger("wikidump.segment.create_dataset")
logger.info("Creating dataset")
docmap = {}
classmap = {}
# Set up a parser for each class
parser = {}
classlabels = []
class_probs = numpy.empty(len(distribution), dtype=float)
dump_indices = []
denominator = 0
logger.info("Setting up parsers")
for i, classlabel in enumerate(distribution):
parser[classlabel] = page_parser(classlabel)
classlabels.append(classlabel)
denominator += distribution[classlabel]
class_probs[i] = denominator
dumpsize = dumpSize(classlabel)
indices = range(dumpsize)
random.shuffle(indices)
dump_indices.append(indices)
logger.debug("Done for %s, %d indices", classlabel, dumpsize)
logger.info("Parsers ready")
# Normalize to a CDF
class_probs /= denominator
random.seed(seed)
logger.info("Obtaining documents")
for doc_num in range(num_doc):
doc = ""
classes = set()
segids = []
for seg_num in range(num_seg):
seg_class_index = numpy.searchsorted(class_probs, random.random())
seg_class = classlabels[seg_class_index]
classes.add(seg_class)
doc_index = dump_indices[seg_class_index].pop(0)
segid = "_".join((seg_class, str(doc_index)))
while segid in exclude:
logger.debug("Ignored %s: in exclude", segid)
try:
doc_index = dump_indices[seg_class_index].pop(0)
except IndexError:
raise ValueError, "No more documents in class %s available" % classlabel
segid = "_".join((seg_class, str(doc_index)))
segids.append(segid)
content = text_transformer(parser[seg_class].get_entry(doc_index))
seg_size = len(content) / num_seg
seg_start = seg_size * seg_num
seg_end = seg_start + seg_size
doc += content[seg_start:seg_end]
docid = "-".join(segids)
docmap[docid] = doc
classmap[docid] = list(classes)
logger.debug("Index: %d ID: %s", doc_num, docid)
return docmap, classmap
while segid in exclude:
logger.debug("Ignored %s: in exclude", segid)
try:
doc_index = dump_indices[seg_class_index].pop(0)
except IndexError:
raise ValueError, "No more documents in class %s available" % classlabel
segid = "_".join((seg_class, str(doc_index)))
segids.append(segid)
content = text_transformer(parser[seg_class].get_entry(doc_index))
seg_size = len(content) / num_seg
seg_start = seg_size * seg_num
seg_end = seg_start + seg_size
doc += content[seg_start:seg_end]
docid = "-".join(segids)
docmap[docid] = doc
classmap[docid] = list(classes)
logger.debug("Index: %d ID: %s", doc_num, docid)
return docmap, classmap
if __name__ == "__main__":
picklefile = open("segment.pickle", "w")
distribution = {}
for p in all_prefixes:
if len(p) > 3:
logger.warning("Ignoring %s", p)
else:
distribution[p] = dumpSize(p)
data = create_dataset(5000, distribution, 2, strip_mediawiki_markup, exclude=[], seed=61383441363)
dump(data, picklefile)