def test_add_parameter():
pName = 'nTopics'
yield assert_false, hasattr(config, pName)
add_parameter(pName, nTopics, jsonFile)
config2 = load_config(jsonFile)
yield assert_equal, config2[pName], nTopics
def setup():
global jsonFile
global config
global nTopics
jsonFile = 'config.json'
# create cofig.json
params = {}
with open(jsonFile, 'wb') as f:
dump(params, f, sort_keys=True, indent=4)
config = load_config(jsonFile)
nTopics = 100
parser = argparse.ArgumentParser()
parser.add_argument('json',
help='json file containing experiment '
'configuration.')
parser.add_argument('--perspectives',
'-p',
help='list of perspectives to '
'calculate contrastive opinions for')
parser.add_argument('--output', '-o', help='file to save output to')
args = parser.parse_args()
print args.perspectives
print args.output
config = load_config(args.json)
if args.output:
fName = args.output
else:
fName = config.get('outDir').format('co_words_{}.csv'.format(
config.get('nTopics')))
logger.info('writing output to {}'.format(fName))
corpus = get_corpus(config)
words = corpus.topic_words()
topics = load_topics(config)
opinions = load_opinions(config)
nks = load_nks(config)
from cptm.utils.experiment import load_config, get_corpus, get_sampler
logger = logging.getLogger(__name__)
logging.basicConfig(format='%(levelname)s : %(message)s', level=logging.INFO)
logging.getLogger('gensim').setLevel(logging.ERROR)
logging.getLogger('CPTCorpus').setLevel(logging.ERROR)
logging.getLogger('CPT_Gibbs').setLevel(logging.ERROR)
parser = argparse.ArgumentParser()
parser.add_argument('json', help='json file containing experiment '
'configuration.')
args = parser.parse_args()
config = load_config(args.json)
corpus = get_corpus(config)
nTopics = config.get('expNumTopics')
nIter = config.get('nIter')
outDir = config.get('outDir')
sampleInterval = 10
for nt in nTopics:
sampler = get_sampler(config, corpus, nTopics=nt, initialize=False)
logging.info('removing parameter sample files for nTopics = {}'.format(nt))
for t in range(sampler.nIter):
if t != 0 and (t+1) % sampleInterval != 0:
try:
remove(sampler.get_theta_file_name(t))
except:
pass
logger = logging.getLogger(__name__)
logging.basicConfig(format='%(levelname)s : %(message)s', level=logging.DEBUG)
logging.getLogger('gensim').setLevel(logging.ERROR)
logging.getLogger('CPTCorpus').setLevel(logging.DEBUG)
logging.getLogger('CPT_Gibbs').setLevel(logging.DEBUG)
parser = argparse.ArgumentParser()
parser.add_argument('json', help='json file containing experiment '
'configuration.')
parser.add_argument('data_dir', help='dir containing the input data.')
parser.add_argument('out_dir', help='dir to write results to.')
args = parser.parse_args()
params = load_config(args.json)
topicDict = params.get('outDir').format('topicDict.dict')
opinionDict = params.get('outDir').format('opinionDict.dict')
c_perspectives = get_corpus(params)
perspectives = [p.name for p in c_perspectives.perspectives]
logger.info('Perspectives found: {}'.format('; '.join(perspectives)))
input_dirs = [args.data_dir for p in perspectives]
corpus = CPTCorpus(input=input_dirs, topicDict=topicDict,
opinionDict=opinionDict, testSplit=100, file_dict=None,
topicLines=params.get('topicLines'),
opinionLines=params.get('opinionLines'))
from CPTCorpus import CPTCorpus
from cptm.utils.experiment import get_sampler, thetaFileName, load_config, \
topicFileName
logger = logging.getLogger(__name__)
logging.basicConfig(format='%(levelname)s : %(message)s', level=logging.DEBUG)
parser = argparse.ArgumentParser()
parser.add_argument('json',
help='json file containing experiment '
'configuration.')
parser.add_argument('data_dir', help='dir containing the input data.')
parser.add_argument('out_dir', help='dir to write results to.')
args = parser.parse_args()
params = load_config(args.json)
input_dir = [args.data_dir]
topicDict = params.get('outDir').format('topicDict.dict')
opinionDict = params.get('outDir').format('opinionDict.dict')
phi_topic_file = topicFileName(params)
phi_topic = pd.read_csv(phi_topic_file, index_col=0, encoding='utf-8').values.T
#print phi_topic.shape
#print phi_topic
corpus = CPTCorpus(input=input_dir,
topicDict=topicDict,
opinionDict=opinionDict,
testSplit=100,
file_dict=None,