gitadmin/foxhunt
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

refactor(ml): delete straggler train_mamba2/train_ppo examples

Browse Source 
These two files survived the 20-file consolidation in 022036cb.
Both are now fully superseded:
- train_ppo.rs → train_baseline_rl --model ppo
- train_mamba2.rs → train_baseline_supervised --model mamba2

Also updates entrypoint-generic.sh usage examples to reference
the unified binaries (train_baseline_rl, train_baseline_supervised).

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
jgrusewski
2026-02-25 23:24:53 +01:00
parent 267240530d
commit 71eefa6d53
3 changed files with 8 additions and 671 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
scripts/entrypoint-generic.sh
View File

@@ -52,17 +52,17 @@ if [ $# -eq 0 ]; then
log " Override Docker CMD with your training command:"
log ""
log "EXAMPLES:"
log " # DQN Training (100 epochs)"
log " CMD: /runpod-volume/binaries/train_dqn --parquet-file /runpod-volume/test_data/ES_FUT_small.parquet --epochs 100 --output-dir /runpod-volume/models"
log " # RL Training (DQN on ES futures, walk-forward)"
log " CMD: /runpod-volume/binaries/train_baseline_rl --model dqn --symbol ES.FUT --data-dir /runpod-volume/data --epochs 100"
log ""
log " # TFT Training (50 epochs, GPU enabled)"
log " CMD: /runpod-volume/binaries/train_tft_parquet --parquet-file /runpod-volume/test_data/ES_FUT_180d.parquet --epochs 50 --use-gpu --output-dir /runpod-volume/models"
log " # RL Training (PPO on NQ futures)"
log " CMD: /runpod-volume/binaries/train_baseline_rl --model ppo --symbol NQ.FUT --data-dir /runpod-volume/data --epochs 200"
log ""
log " # PPO Training (200 epochs)"
log " CMD: /runpod-volume/binaries/train_ppo --parquet-file /runpod-volume/test_data/NQ_FUT_180d.parquet --epochs 200 --output-dir /runpod-volume/models"
log " # Supervised Training (TFT on parquet data)"
log " CMD: /runpod-volume/binaries/train_baseline_supervised --model tft --parquet-file /runpod-volume/test_data/ES_FUT_180d.parquet --epochs 50"
log ""
log " # MAMBA-2 Training (30 epochs)"
log " CMD: /runpod-volume/binaries/train_mamba2_parquet --parquet-file /runpod-volume/test_data/6E_FUT_180d.parquet --epochs 30 --output-dir /runpod-volume/models"
log " # Supervised Training (Mamba2 on parquet data)"
log " CMD: /runpod-volume/binaries/train_baseline_supervised --model mamba2 --parquet-file /runpod-volume/test_data/6E_FUT_180d.parquet --epochs 30"
log ""
log "Container is ready and waiting for commands..."
log "To deploy with a custom command, override the Docker CMD in your deployment script"