metadata

library_name: rl-algo-impls
tags:
  - CarRacing-v0
  - a2c
  - deep-reinforcement-learning
  - reinforcement-learning
model-index:
  - name: a2c
    results:
      - metrics:
          - type: mean_reward
            value: 395.0 +/- 197.79
            name: mean_reward
        task:
          type: reinforcement-learning
          name: reinforcement-learning
        dataset:
          name: CarRacing-v0
          type: CarRacing-v0

A2C Agent playing CarRacing-v0

This is a trained model of a A2C agent playing CarRacing-v0 using the /sgoodfriend/rl-algo-impls repo.

All models trained at this commit can be found at https://api.wandb.ai/links/sgoodfriend/7lx79bf0.

Training Results

This model was trained from 3 trainings of A2C agents using different initial seeds. These agents were trained by checking out 0511de3. The best and last models were kept from each training. This submission has loaded the best models from each training, reevaluates them, and selects the best model from these latest evaluations (mean - std).

algo	env	seed	reward_mean	reward_std	eval_episodes	best	wandb_url
a2c	CarRacing-v0	1	293.812	100.462	16		wandb
a2c	CarRacing-v0	2	395.003	197.795	16	*	wandb
a2c	CarRacing-v0	3	43.5481	43.5812	16		wandb

Prerequisites: Weights & Biases (WandB)

Training and benchmarking assumes you have a Weights & Biases project to upload runs to. By default training goes to a rl-algo-impls project while benchmarks go to rl-algo-impls-benchmarks. During training and benchmarking runs, videos of the best models and the model weights are uploaded to WandB.

Before doing anything below, you'll need to create a wandb account and run wandb login.

Usage

/sgoodfriend/rl-algo-impls: https://github.com/sgoodfriend/rl-algo-impls

Note: While the model state dictionary and hyperaparameters are saved, the latest implementation could be sufficiently different to not be able to reproduce similar results. You might need to checkout the commit the agent was trained on: 0511de3.

# Downloads the model, sets hyperparameters, and runs agent for 3 episodes
python enjoy.py --wandb-run-path=sgoodfriend/rl-algo-impls-benchmarks/i44ri61x

Setup hasn't been completely worked out yet, so you might be best served by using Google Colab starting from the colab_enjoy.ipynb notebook.

Training

If you want the highest chance to reproduce these results, you'll want to checkout the commit the agent was trained on: 0511de3. While training is deterministic, different hardware will give different results.

python train.py --algo a2c --env CarRacing-v0 --seed 2

Setup hasn't been completely worked out yet, so you might be best served by using Google Colab starting from the colab_train.ipynb notebook.

Benchmarking (with Lambda Labs instance)

This and other models from https://api.wandb.ai/links/sgoodfriend/7lx79bf0 were generated by running a script on a Lambda Labs instance. In a Lambda Labs instance terminal:

git clone [email protected]:sgoodfriend/rl-algo-impls.git
cd rl-algo-impls
bash ./lambda_labs/setup.sh
wandb login
bash ./lambda_labs/benchmark.sh [-a {"ppo a2c dqn vpg"}] [-e ENVS] [-j {6}] [-p {rl-algo-impls-benchmarks}] [-s {"1 2 3"}]

Alternative: Google Colab Pro+

As an alternative, colab_benchmark.ipynb, can be used. However, this requires a Google Colab Pro+ subscription and running across 4 separate instances because otherwise running all jobs will exceed the 24-hour limit.

Hyperparameters

This isn't exactly the format of hyperparams in hyperparams/a2c.yml, but instead the Wandb Run Config. However, it's very close and has some additional data:

additional_keys_to_log: []
algo: a2c
algo_hyperparams:
  ent_coef: 1.707983205298309e-07
  gae_lambda: 0.949455676599436
  gamma: 0.99957617037542
  learning_rate: 2.5630993245026736e-05
  learning_rate_decay: linear
  max_grad_norm: 0.5406643389792273
  n_steps: 16
  normalize_advantage: true
  use_rms_prop: false
  vf_coef: 0.10428178193833336
device: auto
env: CarRacing-v0
env_hyperparams:
  frame_stack: 4
  n_envs: 16
  normalize: true
  normalize_kwargs:
    norm_obs: false
    norm_reward: true
env_id: null
eval_params: {}
n_timesteps: 4000000
policy_hyperparams:
  activation_fn: tanh
  cnn_flatten_dim: 256
  hidden_sizes:
  - 256
  init_layers_orthogonal: true
  log_std_init: -1.3502584927786276
  share_features_extractor: false
  use_sde: false
seed: 2
use_deterministic_algorithms: true
wandb_entity: null
wandb_group: null
wandb_project_name: rl-algo-impls-benchmarks
wandb_tags:
- benchmark_0511de3
- host_152-67-249-42
- branch_main
- v0.0.8