Provide more information to the user

[StarCycle]

1. remove abbreviations and use "longer names" in logs
2. enable progress bar for evaluation by default
3. enable asynchronized env by default
4. add readme intro for resuming training

[alexander-soare]

Thanks for these suggestions. I've left some comments.

[StarCycle]

Perhaps you can also print the config at the beginning of training/eval.

Currently you are using hydra and different config files may override each other. Sometimes a user may not remember the setting in another config file, or not know his/her config is override by another config file.

Simply printing all configs at the beginning of a training/eval process can solve this problem, like what they did for mmengine.

[Cadene]

Thanks for your contribution! We will improve onboarding and ease of use.

Left a few comments

[StarCycle]

@Cadene:

Re progressbar: you are right, I will not make it as an option. I still suggest to enable both progress bars (i.e., the progress bar for episodes and the bar for steps in an episode). Users can easily locate problems of evaluation if a step takes too long, or there are too many episodes

[StarCycle]

As a side note, now it logs this at the beginning of training, which is very easy to read:

INFO 2024-08-14 13:15:24       <stdin>:1 {'dataset_repo_id': 'lerobot/pusht',
 'device': 'cuda',
 'env': {'action_dim': 2,
         'episode_length': 300,
         'fps': '${fps}',
         'gym': {'obs_type': 'pixels_agent_pos',
                 'render_mode': 'rgb_array',
                 'visualization_height': 384,
                 'visualization_width': 384},
         'image_size': 96,
         'name': 'pusht',
         'state_dim': 2,
         'task': 'PushT-v0'},
 'eval': {'batch_size': 50, 'n_episodes': 50, 'use_async_envs': True},
 'fps': 10,
 'override_dataset_stats': {'action': {'max': [511.0, 511.0],
                                       'min': [12.0, 25.0]},
                            'observation.image': {'mean': [[[0.5]],
                                                           [[0.5]],
                                                           [[0.5]]],
                                                  'std': [[[0.5]],
                                                          [[0.5]],
                                                          [[0.5]]]},
                            'observation.state': {'max': [496.14618, 510.9579],
                                                  'min': [13.456424,
                                                          32.938293]}},
 'policy': {'beta_end': 0.02,
            'beta_schedule': 'squaredcos_cap_v2',
            'beta_start': 0.0001,
            'clip_sample': True,
            'clip_sample_range': 1.0,
            'crop_is_random': True,
            'crop_shape': [84, 84],
            'diffusion_step_embed_dim': 128,
            'do_mask_loss_for_padding': False,
            'down_dims': [512, 1024, 2048],
            'horizon': 16,
            'input_normalization_modes': {'observation.image': 'mean_std',
                                          'observation.state': 'min_max'},
            'input_shapes': {'observation.image': [3, 96, 96],
                             'observation.state': ['${env.state_dim}']},
            'kernel_size': 5,
            'n_action_steps': 8,
            'n_groups': 8,
            'n_obs_steps': 2,
            'name': 'diffusion',
            'noise_scheduler_type': 'DDPM',
            'num_inference_steps': None,
            'num_train_timesteps': 100,
            'output_normalization_modes': {'action': 'min_max'},
            'output_shapes': {'action': ['${env.action_dim}']},
            'prediction_type': 'epsilon',
            'pretrained_backbone_weights': None,
            'spatial_softmax_num_keypoints': 32,
            'use_film_scale_modulation': True,
            'use_group_norm': True,
            'vision_backbone': 'resnet18'},
 'resume': False,
 'seed': 100000,
 'training': {'adam_betas': [0.95, 0.999],
              'adam_eps': 1e-08,
              'adam_weight_decay': 1e-06,
              'batch_size': 64,
              'delta_timestamps': {'action': '[i / ${fps} for i in range(1 - '
                                             '${policy.n_obs_steps}, 1 - '
                                             '${policy.n_obs_steps} + '
                                             '${policy.horizon})]',
                                   'observation.image': '[i / ${fps} for i in '
                                                        'range(1 - '
                                                        '${policy.n_obs_steps}, '
                                                        '1)]',
                                   'observation.state': '[i / ${fps} for i in '
                                                        'range(1 - '
                                                        '${policy.n_obs_steps}, '
                                                        '1)]'},
              'do_online_rollout_async': False,
              'drop_n_last_frames': 7,
              'eval_freq': 100,
              'grad_clip_norm': 10,
              'image_transforms': {'brightness': {'min_max': [0.8, 1.2],
                                                  'weight': 1},
                                   'contrast': {'min_max': [0.8, 1.2],
                                                'weight': 1},
                                   'enable': False,
                                   'hue': {'min_max': [-0.05, 0.05],
                                           'weight': 1},
                                   'max_num_transforms': 3,
                                   'random_order': False,
                                   'saturation': {'min_max': [0.5, 1.5],
                                                  'weight': 1},
                                   'sharpness': {'min_max': [0.8, 1.2],
                                                 'weight': 1}},
              'log_freq': 200,
              'lr': 0.0001,
              'lr_scheduler': 'cosine',
              'lr_warmup_steps': 500,
              'num_workers': 4,
              'offline_steps': 200000,
              'online_buffer_capacity': None,
              'online_buffer_seed_size': 0,
              'online_env_seed': None,
              'online_rollout_batch_size': 1,
              'online_rollout_n_episodes': 1,
              'online_sampling_ratio': 0.5,
              'online_steps': 0,
              'online_steps_between_rollouts': 1,
              'save_checkpoint': True,
              'save_freq': 100},
 'use_amp': False,
 'video_backend': 'pyav',
 'wandb': {'disable_artifact': False,
           'enable': False,
           'notes': '',
           'project': 'lerobot'}}

[alexander-soare]

Thanks for revising this @StarCycle . My status is now "approving". I will also wait on @Cadene to approve as he has become involved.

Btw, looks like style tests are not passing. Have you seen CONTRIBUTING.md for instructions on how to set up the pre-commit hook?

[Cadene]

@StarCycle By any chance, could you provide code to try this PR?

I feel like at least the third section is missing from the PR description among the sections we advise to add:

1. What the PR adds:
2. How it was tested
3. How to checkout & try? (for the reviewer) <--- example code

See this PR description for instance: #281

Thanks!

[StarCycle]

@Cadene

You are right! I explain it here:

What this does

1. Enable progress bar for evaluation by default, except in slurm
2. Enable asynchronized env by default, except for aloha environments.
3. Add readme intro for resuming training.
4. Add tutorial intro about explainations of abbreviations of the metrics in log
5. It will print all the configuration at the beginning of a training process

How it was tested?

Not too much difference from the original code, just run python lerobot/scripts/train.py policy=diffusion env=pusht

How to checkout and try?

Just run python lerobot/scripts/train.py policy=diffusion env=pusht

[Cadene]

I am very thankful for your time and suggestions.
I left some comments. I think some parts of this PR requires a bit more work.

On our side, we should better convey the purpose for the two sections of the PR description. They are actually quite important to make faster progress on LeRobot. I would have edited the PR description with these info:

How it was tested?

• Ran diffusion training and evaluation on pusht. Configs are displayed and look like this:

• Ran diffusion training and evaluation on pusht. Progress bars are displayed and look like this:

TODO:

• Run training and evaluation on aloha to validate use_async_envs: false
• Run training and evaluation on xarm to validate use_async_envs: false

How to checkout and try?

• To checkout the code:

git remote add starcycle [email protected]:StarCycle/easier_lerobot.git
git fetch starcycle
git checkout starcycle/main

• To run diffusion training and evaluation on pusht:

python lerobot/scripts/train.py policy=diffusion env=pusht \
eval.batch_size=1 eval.use_async_envs=false training.eval_freq=2

• To run diffusion training and evaluation on aloha:
  TODO

• To run diffusion training and evaluation on xarm:
  TODO

Thanks for your help!!!!

[StarCycle]

Just out of curiosity, does LeRobot support multi-gpu training now? (you just mentioned slurm ^^

[Cadene]

@StarCycle Yes we are working on a PR using accelerate: #317

[StarCycle]

@StarCycle Yes we are working on a PR using accelerate: #317

Nice!

[StarCycle]

Hi @Cadene,

Are there other things that I need to complete to merge this PR?

(ﾉ"◑ڡ◑)ﾉ

[Cadene]

Last update required and ready to merge!