SLM Lab
v4.1.1
v4.1.1
  • SLM Lab
  • 🖥Setup
    • Installation
    • Quick Start
  • 🚀Using SLM Lab
    • Lab Command
    • Lab Organization
    • Train and Enjoy: REINFORCE CartPole
    • Agent Spec: DDQN+PER on LunarLander
    • Env Spec: A2C on Pong
    • GPU Usage: PPO on Pong
    • Parallelizing Training: Async SAC on Humanoid
    • Experiment and Search Spec: PPO on Breakout
    • Run Benchmark: A2C on Atari Games
    • Meta Spec: High Level Specifications
    • Post-Hoc Analysis
    • TensorBoard: Visualizing Models and Actions
    • Using SLM Lab In Your Project
  • 📈Analyzing Results
    • Data Locations
    • Graphs and Data
    • Performance Metrics
  • 🥇Benchmark Results
    • Public Benchmark Data
    • Discrete Environment Benchmark
    • Continuous Environment Benchmark
    • Atari Environment Benchmark
    • RL GIFs
  • 🔧Development
    • Modular Design
      • Algorithm Taxonomy
      • Class Inheritance: A2C > PPO
    • Algorithm
      • DQN
      • REINFORCE
      • Actor Critic
    • Memory
      • Replay
      • PrioritizedReplay
      • OnPolicyReplay
      • OnPolicyBatchReplay
    • Net
      • MLP
      • CNN
      • RNN
    • Profiling SLM Lab
  • 📖Publications and Talks
    • Book: Foundations of Deep Reinforcement Learning
    • Talks and Presentations
  • 🤓Resources
    • Deep RL Resources
    • Contributing
    • Motivation
    • Help
    • Contact
Powered by GitBook
On this page

Was this helpful?

  1. 🚀Using SLM Lab

Post-Hoc Analysis

PreviousMeta Spec: High Level SpecificationsNextTensorBoard: Visualizing Models and Actions

Last updated 5 years ago

Was this helpful?

The Retro-Analysis Module

Sometimes we might want to rerun analysis after a Session, Trial or Experiment has completed. For instance, we added new derived metrics or new ways to plotting the graphs. This quick tutorial will show us a command to do so.

Since all the analysis data are generated by the , we can simply rerun it on the Session, Trial or Experiment. This is handled by the .

Let's supposes our data is saved to data/ppo_lunar_2019_11_30_002958 and we wish to run retro_analysis on it. Use the following command:

python -c 'import sys; from slm_lab.experiment import retro_analysis; retro_analysis.retro_analyze("data/ppo_lunar_2019_11_30_002958")'

When invoked, the retro_analysis module will load the data saved at the end of Session, Trial and Experiment, then call the analysis module methods on them. When completed, the new derived data files will overwrite their older counterparts.

The retro_analysis module is safe and free from side-effects – it only overwrites data that is derived, so there is no risk of destroying the data that needs a full rerun to recollect.

🚀
analysis module
retro_analysis module