Skip to content
Snippets Groups Projects
Commit eedd1e00 authored by Franck Dary's avatar Franck Dary
Browse files

Fixed bug in createKFolds

parent 282c818e
No related branches found
No related tags found
No related merge requests found
......@@ -17,6 +17,11 @@ if __name__ == "__main__" :
random.seed(100)
corpusName = sys.argv[1]
while corpusName[-1] == '/' :
corpusName = corpusName[:-1]
corpusName = corpusName.split('/')[-1]
inputFiles = [sys.argv[1]+"/"+filename for filename in os.listdir(sys.argv[1]) if ".conllu" in filename]
sentences = []
......@@ -56,10 +61,11 @@ if __name__ == "__main__" :
train = trainDev[:-testSize]
dev = trainDev[-testSize:]
outDir = sys.argv[2]+"/"+sys.argv[1]
outDir = sys.argv[2]+"/"+corpusName
while outDir[-1] == '/' :
outDir = outDir[:-1]
outDir = outDir + "_" + str(k)
print("Creating '%s'"%outDir, file=sys.stderr)
os.makedirs(outDir, exist_ok=True)
for sents, name in [(train, "train"), (dev, "dev"), (test, "test")] :
with open(outDir + "/" + "%s.conllu"%name, "w") as outFile :
......
0% Loading or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment