Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Add improved data downloading class / pipeline #39

Merged
merged 6 commits into from
Jan 5, 2021
Merged

Add improved data downloading class / pipeline #39

Changes from 1 commit
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
7e3a048
add class for automatically downloading datasets
sdtblck Jan 5, 2021
9cd776f
add class for automatically downloading datasets
sdtblck Jan 5, 2021
07b37d2
cleanup
Jan 5, 2021
f636b2d
add dataset preparation to main train scripts
sdtblck Jan 5, 2021
ee7aad0
fix train data path
sdtblck Jan 5, 2021
a38537e
fix enwik8 errors
sdtblck Jan 5, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
add class for automatically downloading datasets
  • Loading branch information
@sdtblck
sdtblck committed Jan 5, 2021
commit 9cd776ff97d0a714673c0f5b65e1da98eceadc44
4 changes: 2 additions & 2 deletions train_enwik8.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,7 +9,7 @@
from tqdm.auto import trange

from gpt_neox import (GPTNeoX, AutoregressiveWrapper, TextSamplerDataset, download_dataset,
cycle, prepare_optimizer_parameters, decode_tokens, prepare_enwik8_data, is_main)
cycle, prepare_optimizer_parameters, decode_tokens, read_enwik8_data, is_main)


def get_args():
Expand Down Expand Up @@ -46,7 +46,7 @@ def get_params(model):
model = AutoregressiveWrapper(model)

# prepare enwik8 data
data_train, data_val = prepare_enwik8_data(params["data_path"])
data_train, data_val = read_enwik8_data(params["data_path"])
train_dataset = TextSamplerDataset(data_train, params["seq_len"])
val_dataset = TextSamplerDataset(data_val, params["seq_len"])
val_loader = cycle(DataLoader(val_dataset, batch_size=params["batch_size"]))
Expand Down