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

Update checkpointing test for async #1180

Merged
merged 1 commit into from
Jan 24, 2025
Merged

Update checkpointing test for async #1180

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
1 change: 1 addition & 0 deletions MaxText/tests/integration_tests/checkpointing_test.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -34,6 +34,7 @@ def run_checkpointing(attention_type):
"False", # collect_stack_trace
"grain", # dataset_type
attention_type,
"False", # async_checkpointing"
]

subprocess.run(command, check=True, cwd="..")
Expand Down
5 changes: 3 additions & 2 deletions end_to_end/test_checkpointing.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,6 +20,7 @@ ATTENTION=${6}
if [ -z "${6}" ]; then
ATTENTION='autoselected'
fi
ASYNC_CHECKPOINTING=${7:-true}
eval_metrics=checkpoint_save_restore
model_params=" base_emb_dim=384 base_num_query_heads=8 base_num_kv_heads=8 base_mlp_dim=192 base_num_decoder_layers=8 head_dim=128"
CMD_DATA=""
Expand All @@ -37,13 +38,13 @@ fi
#Train
CMD1="python3 MaxText/train.py MaxText/configs/base.yml run_name=$RUN_NAME steps=5 max_target_length=128 per_device_batch_size=1\
metrics_file=saved_metrics.txt checkpoint_period=3 base_output_directory=$OUTPUT_PATH dataset_path=$DATASET_PATH\
async_checkpointing=false collect_stack_trace=$COLLECT_STACK_TRACE attention=$ATTENTION"
async_checkpointing=$ASYNC_CHECKPOINTING collect_stack_trace=$COLLECT_STACK_TRACE attention=$ATTENTION"
CMD1+=$model_params
CMD1+=$CMD_DATA

CMD2="python3 MaxText/train.py MaxText/configs/base.yml run_name=$RUN_NAME steps=5 max_target_length=128 per_device_batch_size=1\
metrics_file=restored_metrics.txt base_output_directory=$OUTPUT_PATH dataset_path=$DATASET_PATH\
async_checkpointing=false collect_stack_trace=$COLLECT_STACK_TRACE attention=$ATTENTION"
async_checkpointing=$ASYNC_CHECKPOINTING collect_stack_trace=$COLLECT_STACK_TRACE attention=$ATTENTION"
CMD2+=$model_params
CMD2+=$CMD_DATA

Expand Down
Loading