Create todset.py
Browse files
todset.py
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
def todset(text: str):
|
2 |
+
lines = [x.rstrip("\n").lower().split("→") for x in text.split("\n")]
|
3 |
+
lines = [(x[0].replace("\\n", "\n"), x[1].replace("\\n", "\n")) for x in lines]
|
4 |
+
|
5 |
+
responses = []
|
6 |
+
for i in lines:
|
7 |
+
if i[1] not in responses:
|
8 |
+
responses.append(i[1])
|
9 |
+
|
10 |
+
dset = {}
|
11 |
+
for sample in tqdm(lines):
|
12 |
+
dset[sample[0]] = responses.index(sample[1])
|
13 |
+
|
14 |
+
return (responses, dset)
|