deep_encoders

TCGBaselines/deep_encoders

deep_encoders is a reinforcement learning model from TCGBaselines. Use it for the reinforcement learning task on the model card, and read the license before you ship it in a product.

This repository contains the complete, checksum-bound JAX checkpoint set referenced by the canonical configs/trainleaderboard.json snapshot.

Downloads · 30 days

Access

Public

Updated Aug 14, 2026

Repo size

10.1 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.gz2 GB · 100%

At a glance

Task: Reinforcement Learning
Access: Public
Created: Aug 9, 2026
Updated: Aug 14, 2026
SHA: 4a10e167

Model card

From the Hugging Face model README

Current deep specialist encoder leaderboard

This repository contains the complete, checksum-bound JAX checkpoint set referenced by the canonical configs/train_leaderboard.json snapshot.

Source code Git commit: 1d9319dcd7ee2154ec239f74c12cfba1e4f68c75
Packaged worktree board Git blob: 40c01441600baaedc909affc8a88f8b10d40c128
Board blob at source commit: 40c01441600baaedc909affc8a88f8b10d40c128
Packaged board matches source commit: true
Source board SHA-256: e48b67dbc74216a1b8f401f5ddced303f1828f1f0b0c2d46d33ad4e0e69fe822
Board members / unique deck hashes: 26 / 26
Unique checkpoint directories: 26
Direct deck-trained members: 26

Every payload is a complete entity_interaction_deep or entity_interaction_deep_v2 stateless_mlp, specialist_rule_state_v1, public-event-window-0 checkpoint. The archive retains every file in each referenced checkpoint directory and deduplicates directories reused by more than one board member.

Download, verify, and extract

hf download TCGBaselines/deep_encoders \
  deep_encoders_train_leaderboard_current.tar.gz deep_encoders_train_leaderboard_current.tar.gz.sha256 --local-dir .
sha256sum -c deep_encoders_train_leaderboard_current.tar.gz.sha256
tar -xzf deep_encoders_train_leaderboard_current.tar.gz
sha256sum -c deep_encoders_bundle_checksums.sha256

Extract at the ladderboard_elo repository root. This creates the exact checkpoints/<id>/ paths plus configs/train_leaderboard.json, MODELS.json, DEEP_ENCODERS_BUNDLE_README.md, and the internal checksum manifest. Existing files at those paths are replaced by normal tar extraction, so use a clean checkout or back up local changes first.

Cross-deck actor reuse

board_deck_hash and checkpoint_trained_deck_hash are separate fields in MODELS.json. The following entries intentionally run an actor on a target deck other than the deck recorded by its immutable checkpoint manifest:

Member	Board deck	Checkpoint-trained deck	Relation

These are valid frozen actor opponents, but their stored critics are not declared compatible with the retargeted deck. Use actor-only initialization and a fresh/warm-started target-deck critic unless newer deck-specific evidence explicitly says otherwise. The immutable checkpoint manifests are preserved rather than rewritten to conceal training provenance.

Current board members

Rank	Elo	Member	Board deck	Checkpoint relation
1	1048.6	`lopbunny_dudunsparce_phase2_u000443_43b70a`	`04689d80f09215b6`	`direct`
2	999.9	`cornerstone_ogerpon_crustle_mega_kangaskhan_manual_v2opp_even_u000100_7046cc`	`e2194d835b594811`	`direct`
3	959.1	`mega_lucario_phase3_best_u000200_c57fbe`	`2796018749350c2b`	`direct`
4	946.2	`deep_mega_starmie_cinderace_phase2_u000196_db88e4`	`00bbd953f688a18b`	`direct`
5	945.2	`mega_lopunny_froslass_phase3_best_u000240_e66e1f`	`07818719c2c434ef`	`direct`
6	942.7	`blaziken_dragapult_deep_vrpo_latest_u000420_966f71`	`362eca0b8a31393e`	`direct`
7	942.0	`archaludon_cinderace_phase3_best_u000240_107c09`	`82ce67869190aefd`	`direct`
8	940.7	`festival_lead_manual_v2opp_even_u000200_0afdbe`	`07d4c9cb1496190b`	`direct`
9	940.3	`hop_trevenant_dudunsparce_deep_vrpo_final_u000740`	`850f423d1fe52b95`	`direct`
10	938.9	`tr_honchkrow_porygon2_phase3_best_u000280_427778`	`d71c667d3ed25c89`	`direct`
11	937.3	`alakazam_dudunsparce_phase3_best_u000160_96ab07`	`01dc0c8ee2afb65c`	`direct`
12	927.3	`starmie_froslass_phase2_u000925_efff34`	`0e6ead751f2c6e11`	`direct`
13	927.0	`tr_mewtwo_spidops_deep_vrpo_u000480_620d59`	`ec38a9f1af1ad0f3`	`direct`
14	919.9	`dragapult_dusknoir_phase1_u000220_8e22f9`	`f9f5dc7482675db1`	`direct`
15	916.8	`lucario_dunsparce_deep_vrpo_continuation_u000760_1c1db6`	`8e33dea1649f0c39`	`direct`
16	913.2	`crustle_kangaskhan_deep_ppo_phase2_u000121_b43489`	`2959276a9ee66220`	`direct`
17	893.4	`prague_dragapult_exact_cont_u000190_15a109`	`079d6dfeb312242b`	`direct`
18	877.1	`grimmsnarl_froslass_phase3_best_u000280_c63b55`	`2a9b67c9561fbdd4`	`direct`
19	876.1	`clefairy_hammer_dragapult_manual_v2opp_u000310_faa5aa`	`c63916cd72ec8788`	`direct`
20	865.4	`festival_lead_arian_improved_manual_v2opp_u000210_b67ce8`	`eb27f50f2a0327ea`	`direct`
21	862.3	`cynthia_garchomp_phase2_u000200_bbbade`	`7218e9812dd0ef05`	`direct`
22	853.1	`dragapult_dusknoir_cont_u000130_ccff55`	`a9358b44a6c7aa2e`	`direct`
23	816.2	`mega_abomasnow_kyogre_deep_vrpo_u000400_774a80`	`b5a0ec95d155c444`	`direct`
24	800.0	`raging_bolt_ba07_phase1_best_u000300_e5f767`	`ba07b48dc438899f`	`direct`
25	742.5	`hydrapple_ogerpon_phase3_best_u000280_726cff`	`efe238b2a77e2d1f`	`direct`
26	0.0	`okidogi_barbaracle_deep_vrpo_latest_u000280_d7bc4d`	`b15fcfd19acc20ed`	`direct`

These are game-playing specialist policies, not Transformers models or a hosted inference endpoint. Elo calibration and provenance remain in the shipped board and MODELS.json.

Task: Reinforcement Learning
Created: Aug 9, 2026
Updated: Aug 14, 2026