This task involves implementing subword tokenization using the BPE algorithm. **Example:** The input 'unbreakable' should return ['un', '##break', '##able']. **Constraints:** The input will be a string, and the output should be a list of strings.
Test Cases
Test Case 1
Input:
'unbreakable'Expected:
['un', '##break', '##able']Test Case 2
Input:
'unbelievable'Expected:
['un', '##be', '##liev', '##able']+ 3 hidden test cases