train model
Browse files- scripts/train_model.py +1 -1
scripts/train_model.py
CHANGED
@@ -225,7 +225,7 @@ training_args = TrainingArguments(
|
|
225 |
per_device_train_batch_size=1, # Adjust based on your GPU memory
|
226 |
per_device_eval_batch_size=1,
|
227 |
optim='adamw_bnb_8bit',
|
228 |
-
gradient_accumulation_steps=
|
229 |
gradient_checkpointing=True,
|
230 |
warmup_steps=500,
|
231 |
weight_decay=0.01,
|
|
|
225 |
per_device_train_batch_size=1, # Adjust based on your GPU memory
|
226 |
per_device_eval_batch_size=1,
|
227 |
optim='adamw_bnb_8bit',
|
228 |
+
gradient_accumulation_steps=8,
|
229 |
gradient_checkpointing=True,
|
230 |
warmup_steps=500,
|
231 |
weight_decay=0.01,
|