Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / hprairie/snake_reinforcement_learning issues and pull requests
#12 - Add prioritized memory replacement
Issue -
State: open - Opened by Hprairie about 1 year ago
#11 - Change Prioritized Replay Experience to not sample from entire distribution
Issue -
State: open - Opened by Hprairie about 1 year ago
#10 - Fix model saving for deuling architecture in DQN and DDQN
Issue -
State: open - Opened by Hprairie about 1 year ago
#9 - Fix Prioritized Double DQN Agent
Issue -
State: closed - Opened by Hprairie about 1 year ago
#8 - Refactor Buffer Object, allowing for more customization
Issue -
State: closed - Opened by Hprairie over 1 year ago
#7 - Fix model saving to save both target and online model
Issue -
State: closed - Opened by Hprairie over 1 year ago
#6 - Added exploration strategy information to saving network
Issue -
State: closed - Opened by Hprairie over 1 year ago
#5 - Add utils for displaying and storing logging information
Issue -
State: closed - Opened by Hprairie over 1 year ago
#4 - Add loading to DQN agent enabling restarting training
Issue -
State: closed - Opened by Hprairie over 1 year ago
#3 - Add start_length and display_game functionality to SnakeGame
Issue -
State: open - Opened by Hprairie over 1 year ago
#2 - Add seeding to randomness throughout the project
Issue -
State: closed - Opened by Hprairie over 1 year ago
#1 - Add Comments to model.py and agents.py
Issue -
State: open - Opened by Hprairie over 1 year ago