library_name: rl-algo-impls
tags:
- Walker2DBulletEnv-v0
- a2c
- deep-reinforcement-learning
- reinforcement-learning
model-index:
- name: a2c
results:
- metrics:
- type: mean_reward
value: 912.59 +/- 246.85
name: mean_reward
task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: Walker2DBulletEnv-v0
type: Walker2DBulletEnv-v0
A2C Agent playing Walker2DBulletEnv-v0
This is a trained model of a A2C agent playing Walker2DBulletEnv-v0 using the /sgoodfriend/rl-algo-impls repo.
All models trained at this commit can be found at https://api.wandb.ai/links/sgoodfriend/eyvb72mv.
Training Results
This model was trained from 3 trainings of A2C agents using different initial seeds. These agents were trained by checking out 0760ef7. 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 | Walker2DBulletEnv-v0 | 1 | 786.406 | 227.244 | 12 | wandb | |
| a2c | Walker2DBulletEnv-v0 | 2 | 912.595 | 246.854 | 12 | * | wandb |
| a2c | Walker2DBulletEnv-v0 | 3 | 427.068 | 419.115 | 12 | 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: 0760ef7.
# Downloads the model, sets hyperparameters, and runs agent for 3 episodes
python enjoy.py --wandb-run-path=sgoodfriend/rl-algo-impls-benchmarks/z53dyb7q
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: 0760ef7. While training is deterministic, different hardware will give different results.
python train.py --algo a2c --env Walker2DBulletEnv-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/eyvb72mv were generated by running a script on a Lambda Labs instance. In a Lambda Labs instance terminal:
git clone git@github.com:sgoodfriend/rl-algo-impls.git
cd rl-algo-impls
bash ./lambda_labs/setup.sh
wandb login
bash ./lambda_labs/benchmark.sh
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:
algo: a2c
algo_hyperaparms:
ent_coef: 0
gae_lambda: 0.9
gamma: 0.99
learning_rate: 0.00096
learning_rate_decay: linear
max_grad_norm: 0.5
n_steps: 8
vf_coef: 0.4
env: Walker2DBulletEnv-v0
env_hyperparams:
n_envs: 4
normalize: true
n_timesteps: 2000000
policy_hyperparams:
init_layers_orthogonal: false
log_std_init: -2
use_sde: true
seed: 2
use_deterministic_algorithms: true
wandb_entity: null
wandb_project_name: rl-algo-impls-benchmarks
wandb_tags:
- benchmark_0760ef7
- host_192-9-248-209