Search results

Finetuning Calql for envs > 1?

Status: Open.

#59 In irom-princeton/dppo;

· yiqiwang8177 opened on Mar 15, 2026

Irreproducible result

Status: Open.

#58 In irom-princeton/dppo;

· hongsukchoi opened on Oct 15, 2025

A technically basic question

Status: Open.

#57 In irom-princeton/dppo;

· he-nantian opened on Sep 4, 2025

0 Improvement after 8M env steps finetuning on Transport

Status: Open.

#56 In irom-princeton/dppo;

· wkguo opened on Aug 20, 2025

Question about single env

Status: Open.

#55 In irom-princeton/dppo;

· lulu6868 opened on Aug 3, 2025

Expected input to have 3 channels, but got 96 channels instead during eval_diffusion_mlp_img run

Status: Open.

#54 In irom-princeton/dppo;

· Ju6276 opened on Jul 27, 2025

Inability to reproduce Section 5.2, comparison to other demo-augmented RL algorithms

Status: Open.

#53 In irom-princeton/dppo;

· hetankevin opened on Jul 25, 2025

Eval success rate is 0, but task appears successful in visualization (avoid_m1)

Status: Open.

#52 In irom-princeton/dppo;

· Ju6276 opened on Jul 7, 2025

Always return done = False, is it reasonable?

Status: Open.

#51 In irom-princeton/dppo;

· Rooshy-yang opened on Jun 30, 2025

Question about BC loss

Status: Open.

#50 In irom-princeton/dppo;

· Doctor-James opened on May 19, 2025

Encountered a bug while running the model evaluation script.

Status: Open.

#49 In irom-princeton/dppo;

· HanwenWangBUPT opened on May 15, 2025

How to calculate log_prob when use_ddim = True?

Status: Open.

#48 In irom-princeton/dppo;

· chenkang455 opened on May 1, 2025