tacit/nanochat
1
0
Fork 0
mirror of https://github.com/karpathy/nanochat.git synced 2025-12-06 04:12:13 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Merge 26b0941f75 into 4a87a0d19f

Browse Source
This commit is contained in:
Pyry Takala 2025-11-22 22:51:48 +01:00 committed by GitHub
commit 0fb04c8f25
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 12 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
tasks/common.py
View File

@ -6,6 +6,9 @@ Example tasks: MMLU, ARC-Easy, ARC-Challenge, GSM8K, HumanEval, SmolTalk.
""" """
import random import random
import logging
logger = logging.getLogger(__name__)
class Task: class Task:
""" """
@ -34,7 +37,15 @@ class Task:
def __len__(self): def __len__(self):
start = self.start start = self.start
stop = self.num_examples() if self.stop is None else self.stop num_ex = self.num_examples()
if self.stop is not None and self.stop > num_ex:
# Warn once, then cap stop
logger.warning(
f"Stop parameter ({self.stop}) exceeds dataset size ({num_ex}). "
f"Using {num_ex} examples instead."
)
self.stop = num_ex
stop = num_ex if self.stop is None else self.stop
step = self.step step = self.step
span = stop - start span = stop - start
num = (span + step - 1) // step # ceil_div(span, step) num = (span + step - 1) // step # ceil_div(span, step)