Create phi_tokenizer.py
Browse files- phi_tokenizer.py +23 -0
phi_tokenizer.py
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
class Tokenizer:
|
| 2 |
+
def __init__(self):
|
| 3 |
+
self.tokens = []
|
| 4 |
+
|
| 5 |
+
def tokenize(self, text):
|
| 6 |
+
for char in text:
|
| 7 |
+
self.tokens.append(ord(char))
|
| 8 |
+
return self.tokens
|
| 9 |
+
|
| 10 |
+
def detokenize(self, tokens):
|
| 11 |
+
text = ''.join(chr(token) for token in tokens)
|
| 12 |
+
return text
|
| 13 |
+
|
| 14 |
+
def main():
|
| 15 |
+
phi_tokenizer = Tokenizer()
|
| 16 |
+
text = "hello world"
|
| 17 |
+
tokenized = phi_tokenizer.tokenize(text)
|
| 18 |
+
print(tokenized)
|
| 19 |
+
detokenized = phi_tokenizer.detokenize(tokenized)
|
| 20 |
+
print(detokenized)
|
| 21 |
+
|
| 22 |
+
if __name__ == "__main__":
|
| 23 |
+
main()
|