• File size measurements show the distribution of size of files.
• Files are classified in four categories based on their size (lines of code): 1-100 (very small files), 101-200 (small files), 201-500 (medium size files), 501-1000 (long files), 1001+(very long files).
• It is a good practice to keep files small. Long files may become "bloaters", code that have increased to such gargantuan proportions that they are hard to work with.

• There are 20 files with 7,078 lines of code.
• 1 very long files (1,303 lines of code)
• 4 long files (2,778 lines of code)
• 8 medium size files (2,475 lines of codeclsfd_ftr_w_mp_ins)
• 2 small files (339 lines of code)
• 5 very small files (183 lines of code)

There are 12 files with lines longer than 120 characters. In total, there are 161 long lines.

File	# lines	# units	# long lines
pytorch_ddp.ipynb in pai-python-sdk/training/pytorch_ddp	557	-	92
huggingface_bert.ipynb in pai-python-sdk/huggingface/huggingface_bert	865	-	17
xgboost_breast_cancer.ipynb in pai-python-sdk/framework/xgboost_breast_cancer	832	-	12
checkpoint.ipynb in pai-python-sdk/training/checkpoint	1303	-	11
pretrained_model.ipynb in pai-python-sdk/training/pretrained_model	524	-	6
train_with_experiment.ipynb in pai-python-sdk/training/train_with_experiment	393	-	5
modelscope_model_deploy.ipynb in pai-python-sdk/modelscope/modelscope_model_deploy	230	-	4
async_inference.ipynb in pai-python-sdk/inference/async_inference	479	-	4
tensorboard.ipynb in pai-python-sdk/training/tensorboard	261	-	4
huggingface_model_deploy.ipynb in pai-python-sdk/huggingface/huggingface_model_deploy	195	-	3
model_deploy_container.ipynb in pai-python-sdk/inference/model_deploy_container	392	-	2
openai_client.py in model-gallery/deploy/mllm	76	3	1