← All tasks
pythonLatentActionPretraining/LAPA #36Not a task: not reproduced

environment configuration

envgap__LatentActionPretraining__LAPA-36

Original GitHub issue ↗Opened 2025-04-03

01 / FAILURE SIGNATURE

As reported upstream

At this point, I get the following error:
Not a benchmark task.
  • No curated issue-specific recipe or verified environment fix is available.

02 / ENVIRONMENT RECIPE

Base commit
Not freshly verified
Manifest
requirements.txt
Reproduce
Awaiting issue-specific recipe
Run under trace
Awaiting a meaningful runtime command

03 / ORIGINAL ISSUE TEXT

LatentActionPretraining/LAPA #36 · read the original issue
This is my complete process of configuring the environment, I hope it can be helpful to you
### 1. programming environment setup with cuda: 12.2

```
conda create -n lapaTemp python=3.10 -y
conda activate lapaTemp

```
Then clone the code:
`git clone https://github.com/LatentActionPretraining/LAPA.git
`
Remove the jax installation from the original project. The new documents `requirement.txt` are as follows:
```
git+https://github.com/lhao499/tux.git
flax==0.7.0
optax==0.1.7
chex==0.1.82
einops
transformers==4.29.2
datasets==2.13.0
tqdm
ml_collections
wandb
gcsfs
requests
typing-extensions
sentencepiece
Pillow
ipdb
imageio[ffmpeg]
decord
tiktoken
tensorflow[and-cuda]
scipy==1.12.0
albumentations
uvicorn
fastapi
```
execution instruction
```pip install -r requirements.txt ```
Then installation jax/jaxlib:
```pip install --upgrade jax==0.4.23 jaxlib==0.4.23+cuda12.cudnn89 -f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html```
At this point, I get the following error:

> (lapaTemp) lyd@zkyd:/data/lyd/workspace/LAPA$ pip install --upgrade jax==0.4.23 jaxlib==0.4.23+cuda12.cudnn89 -f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html
> Looking in indexes: https://pypi.tuna.tsinghua.edu.cn/simple
> Looking in links: https://storage.googleapis.com/jax-releases/jax_cuda_releases.html
> Collecting jax==0.4.23
>   Using cached https://pypi.tuna.tsinghua.edu.cn/packages/28/d0/edf653ea02628f2130ea2557f96d02b264768a2f54d22a9c002c7119cb1d/jax-0.4.23-py3-none-any.whl (1.7 MB)
> Collecting jaxlib==0.4.23+cuda12.cudnn89
>   Using cached https://storage.googleapis.com/jax-releases/cuda12/jaxlib-0.4.23%2Bcuda12.cudnn89-cp310-cp310-manylinux2014_x86_64.whl (131.8 MB)
> Requirement already satisfied: ml-dtypes>=0.2.0 in /data/lyd/miniconda3/envs/lapaTemp/lib/python3.10/site-packages (from jax==0.4.23) (0.5.1)
> Requirement already satisfied: numpy>=1.22 in /data/lyd/miniconda3/envs/lapaTemp/lib/python3.10/site-packages (from jax==0.4.23) (1.26.4)
> Requirement already satisfied: opt-einsum in /data/lyd/miniconda3/envs/lapaTemp/lib/python3.10/site-packages (from jax==0.4.23) (3.4.0)
> Requirement already satisfied: scipy>=1.9 in /data/lyd/miniconda3/envs/lapaTemp/lib/python3.10/site-packages (from jax==0.4.23) (1.12.0)
> Installing collected packages: jaxlib, jax
>   Attempting uninstall: jaxlib
>     Found existing installation: jaxlib 0.5.3
>     Uninstalling jaxlib-0.5.3:
>       Successfully uninstalled jaxlib-0.5.3
>   Attempting uninstall: jax
>     Found existing installation: jax 0.5.3
>     Uninstalling jax-0.5.3:
>       Successfully uninstalled jax-0.5.3
> ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
> orbax-checkpoint 0.11.10 requires jax>=0.5.0, but you have jax 0.4.23 which is incompatible.
> Successfully installed jax-0.4.23 jaxlib-0.4.23+cuda12.cudnn89


The orbax-checkpoint 0.11.10 version 
Continue on GitHub ↗

04 / LABELS

Labels from the report text only; not yet run

No supported category has been assigned.

Label rules and the text that matched
[]

Issue-specific recipe and runtime smoke command require review against the complete issue and repository.

Legacy already_works is install-only; proposed control still requires runtime verification.