def main():
"""
generate dataset
"""
data_path = "../../../data/"
# set up container for data
tax = taxonomy.create_tax_four()
#tax = taxonomy.create_tax_many()
dh = DataHandler(tax)
for org in tax.get_data_keys():
org_id = dh.task_to_id[org]
print "processing organism %s (id=%i)" % (org, org_id)
org_path = data_path + org + "/"
save_path = org_path
ptt_files = [ptt for ptt in os.listdir(org_path) if ptt.endswith(".ptt")]
print("Gff files: %s",ptt_files)
ptt_file_sizes = [os.stat(org_path + ptt).st_size for ptt in ptt_files]
print ptt_file_sizes
# pick largest one (in case we have several contigs)
largest_idx = 0;
if (len(ptt_file_sizes)>1):
print 'There are multiple .ptt files available for this organism.'
print 'I only use the biggest one.'
largest_idx = numpy.argmax(ptt_file_sizes)
print 'Largest file is: ', largest_idx
# determine file names
selected_ptt = org_path + ptt_files[largest_idx]
selected_fna = selected_ptt.replace(".ptt", ".fna")
# invoke generation procedure
dh.add_organism(org_id, selected_fna, selected_ptt)
import ipdb
ipdb.set_trace()
# store final result
save_path = "/tmp/"
dh.save_to_file(save_path)
def main():
tax = taxonomy.create_tax_four()
train(tax)