osmarks/nanogpt-experiments
1
0
Fork 0
mirror of https://github.com/osmarks/nanogpt-experiments.git synced 2025-10-31 15:23:01 +00:00
Code Issues Releases Wiki Activity

Merge pull request #301 from okuvshynov/master

Browse Source 
[easy] allow multithreading in load_dataset
This commit is contained in:
Andrej
2023-06-16 18:30:03 -07:00
committed by GitHub
parent 7339b904ef bb7e96754a
commit 41d7014f7d
1 changed files with 6 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
data/openwebtext/prepare.py
View File

@@ -11,8 +11,13 @@ from datasets import load_dataset # huggingface datasets
# good number to use is ~order number of cpu cores // 2 # good number to use is ~order number of cpu cores // 2
num_proc = 8 num_proc = 8
# number of workers in load_dataset() call
# best number might be different from num_proc above as it also depends on NW speed.
# it is better than 1 usually though
num_proc_load_dataset = num_proc
# takes 54GB in huggingface .cache dir, about 8M documents (8,013,769) # takes 54GB in huggingface .cache dir, about 8M documents (8,013,769)
dataset = load_dataset("openwebtext") dataset = load_dataset("openwebtext", num_proc=num_proc_load_dataset)
# owt by default only contains the 'train' split, so create a test split # owt by default only contains the 'train' split, so create a test split
split_dataset = dataset["train"].train_test_split(test_size=0.0005, seed=2357, shuffle=True) split_dataset = dataset["train"].train_test_split(test_size=0.0005, seed=2357, shuffle=True)