flax-community
/

gpt2-medium-indonesian

Text Generation

text-generation-inference

Inference Endpoints

Model card Files Files and versions Metrics Training metrics Community

cahya commited on Jul 7, 2021

Commit

a9c5d2c

•

1 Parent(s): 60f90b3

reduced the bs

Files changed (1) hide show

run_pretraining.sh +2 -2

run_pretraining.sh CHANGED Viewed

@@ -11,8 +11,8 @@ export WANDB_LOG_MODEL="true"
     --dataset_config_name="unshuffled_deduplicated_id" \
     --do_train --do_eval \
     --block_size="512" \
-    --per_device_train_batch_size="64" \
-    --per_device_eval_batch_size="64" \
     --learning_rate="5e-3" --warmup_steps="1000" \
     --adam_beta1="0.9" --adam_beta2="0.98" --weight_decay="0.01" \
     --overwrite_output_dir \

     --dataset_config_name="unshuffled_deduplicated_id" \
     --do_train --do_eval \
     --block_size="512" \
+    --per_device_train_batch_size="24" \
+    --per_device_eval_batch_size="24" \
     --learning_rate="5e-3" --warmup_steps="1000" \
     --adam_beta1="0.9" --adam_beta2="0.98" --weight_decay="0.01" \
     --overwrite_output_dir \