Checkpoint User Guide


In many cases, Stand-alone training and Distributed training can be aborted by the software problem or hardware problem. More seriously, we waste so much time and the performance of the machine but get nothing, which makes us frustrating and we have to restart it again.


The feature of Checkpoint can save Intermediate model variables, lookup table variable, and other needs data in checkpoint directory. When the exception occurs, we can load these variables from the checkpoint directory immediately.


Complete Features Currently:

  1. The Trainer 0 will save model variables in training.

  2. Each of the Trainer will save its own arguments needed.

  3. Each of the Parameter Server will save Distribute Lookup Table variables in training.

Fluid Checkpoint directory structure:

checkpoint_dir (the checkpoint directory user define)
├── checkpoint_0 (the first save directory)
│   ├── __lockup_table__ (Distribute Lookup Table directory)
│   │   ├── table_pserver_0 (Lookup table's data about Pserver 0)
│   │   └── table_pserver_1
│   ├── __model__ (model directory)
│   │   └── var.w_1
│   └── trainer_0 (each trainer will save its own data)
│       ├── epoch_id
│       └── step_id
└── checkpoint_1 (the second save directory)


Fluid.CheckpointConfig construct

When the user wants to use Checkpoint feature, the main thing user have to do is declare CheckpointConfig and construct it.

CheckpointConfig has 4 member variables need to be initialized:

| Member Variable | Type | Comment | | - | :-: | - | | checkpoint_dir | int| checkpoint directory | | max_num_checkpoints | int | Maximum number of checkpoint copies | | epoch_interval | int | epoch interval times | | step_interval | int | step interval times |

Add Fluid.CheckpointConfig’s declaration in Fluid.Trainer

Because the initialization of Trainer needs an instance of CheckpointConfig., we should declare CheckpointConfig in Fluid first.

For example:

config = CheckpointConfig(
    checkpoint_dir = "/tmp/ckpt", max_num_checkpoints = 2, 
    epoch_interval = 2, step_interval = 10)
trainer = Trainer(..., checkpoint_config=config)

After all the things done, the train will save checkpoint at the specified epoch and step, when the train is aborted, the user can restart it, the train will restore from the latest copy.


  1. Make the checkpoint_dir only be used by one train job.

  2. The number of max_num_checkpoints need to be adjusted by the disk size and model size.

  3. Too frequently to slow down the train speed, so too small epoch_interval and step_interval are not suitable.

  4. In distributed train, each Trainer will save arguments in its checkpoint_dir (Only Trainer 0 will save model variables). We need distributed file system (HDFS, etc) to merge all the checkpoint_dir to get the whole data.