Metamon Parsed Replays: Pokémon Showdown Gameplay Trajectories

Metamon Parsed Replays is a dataset of Pokémon Showdown replay files parsed into reinforcement learning trajectories. The dataset was created by author jakegrigsby and was last updated on Hugging Face on May 21, 2026. It is intended for use with the Metamon library's dataloader.

Use Cases

• Training reinforcement learning agents based on parsed human gameplay trajectories.
• Benchmarking agent performance against human strategies based on reconstructed replay data.
• Analyzing game state transitions and player decision-making based on the RL trajectory format.

Strengths

• Data is specifically parsed for reinforcement learning applications, as referenced in the description.
• Includes a dedicated dataloader interface (metamon.data.ParsedReplayDataset) for integration.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.

Provenance

Source

jakegrigsby on Hugging Face

Collection Method

Parsed or reconstructed from Pokémon Showdown replay files.

Freshness

Last updated 2026-05-21 05:28:09; freshness should be verified.