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

Train log gcs #36

Merged
merged 5 commits into from
Jun 25, 2024
Merged

Train log gcs #36

Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
a4ce96c
add gcs upload for composer train
lruizcalico Jun 19, 2024
7116151
fix folder name
lruizcalico Jun 20, 2024
6f4ad56
Merge remote-tracking branch 'origin/main' into train-log-gcs
lruizcalico Jun 21, 2024
e4ce9af
upload train valid folders
lruizcalico Jun 21, 2024
4acf8a2
fix folder
lruizcalico Jun 21, 2024
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
21 changes: 20 additions & 1 deletion src/baskerville/trainer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,7 +17,8 @@

import numpy as np
import tensorflow as tf

import tempfile
from baskerville.helpers.gcs_utils import is_gcs_path, upload_folder_gcs
from baskerville import metrics


Expand Down Expand Up @@ -119,6 +120,15 @@ def __init__(
self.batch_size = self.train_data[0].batch_size
self.compiled = False

# if log_dir is in gcs then create a local temp dir
if is_gcs_path(self.log_dir):
folder_name = "/".join(self.log_dir.split("/")[3:])
self.log_dir = tempfile.mkdtemp() + "/" + folder_name
self.gcs_log_dir = log_dir
self.gcs = True
else:
self.gcs = False

# early stopping
self.patience = self.params.get("patience", 20)

Expand Down Expand Up @@ -498,6 +508,10 @@ def eval_step1_distr(xd, yd):
print(" - valid_r2: %.4f" % valid_r2[di].result().numpy(), end="")
early_stop_stat = valid_r[di].result().numpy()

# upload to gcs
if self.gcs:
upload_folder_gcs(train_log_dir, self.gcs_log_dir)
upload_folder_gcs(valid_log_dir, self.gcs_log_dir)
# checkpoint
managers[di].save()
model.save(
Expand Down Expand Up @@ -697,6 +711,11 @@ def eval_step_distr(xd, yd):
end="",
)

# upload to gcs
if self.gcs:
upload_folder_gcs(train_log_dir, self.gcs_log_dir)
upload_folder_gcs(valid_log_dir, self.gcs_log_dir)

# checkpoint
manager.save()
seqnn_model.save("%s/model_check.h5" % self.out_dir)
Expand Down
Loading