Fix several syntax warning related to invalid escape sequence (#144)

by using raw strings or properly escaping the backslashes.

```log
AReaL/realhf/impl/dataset/math_parser.py:292: SyntaxWarning: invalid escape sequence '\%'
  string = string.replace("\%", "")
AReaL/realhf/impl/dataset/math_parser.py:402: SyntaxWarning: invalid escape sequence '\d'
  pattern = "-?\d*\.?\d+"
AReaL/realhf/impl/model/parallelism/tensor_parallel/modules.py:1125: SyntaxWarning: invalid escape sequence '\s'
```

Signed-off-by: Hollow Man <hollowman@opensuse.org>
This commit is contained in:
ℍ𝕠𝕝𝕝𝕠𝕨 𝕄𝕒𝕟 2025-07-01 05:12:39 +03:00 committed by GitHub
parent 254614df87
commit e0aee03109
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 2 additions and 3 deletions

View File

@ -289,7 +289,6 @@ def strip_string(string, skip_unit=False):
# remove percentage
string = string.replace("\\%", "")
string = string.replace("\%", "")
string = string.replace("%", "")
# " 0." equivalent to " ." and "{0." equivalent to "{." Alternatively, add "0" if "." is the start of the string
@ -399,7 +398,7 @@ def extract_answer(pred_str, data_name, use_last_number=True):
pred = pred_str.split("答案是")[1].strip().split("\n\n")[0].strip()
else: # use the last number
if use_last_number:
pattern = "-?\d*\.?\d+"
pattern = r"-?\d*\.?\d+"
pred = re.findall(pattern, pred_str.replace(",", ""))
if len(pred) >= 1:
pred = pred[-1]

View File

@ -1122,7 +1122,7 @@ class _VocabParallelCrossEntropy(torch.autograd.Function):
vocab_size = exp_logits.size(-1)
if label_smoothing > 0:
"""
r"""
We'd like to assign 1 / (K - 1) probability mass to every index that is not the ground truth.
= (1 - alpha) * y_gt + alpha * mean(y_{i for i != gt})
= (1 - alpha) * y_gt + (alpha / (K - 1)) * \sum_{i != gt} y_i