Skip to content

fix: fix dockerfile for changes to vllm and transformers#229

Merged
pjannaty merged 7 commits into
mainfrom
rvaibhav/issue_210_vllm
Oct 29, 2025
Merged

fix: fix dockerfile for changes to vllm and transformers#229
pjannaty merged 7 commits into
mainfrom
rvaibhav/issue_210_vllm

Conversation

@rvaibhav

@rvaibhav rvaibhav commented Oct 24, 2025 •
edited
Loading

Copy link
Copy Markdown
Collaborator

Fix for issue #210

@rvaibhav
rvaibhav requested review from codeJRV and nhayesroth and removed request for codeJRV October 28, 2025 16:22
@pjannaty
pjannaty requested review from spectralflight and removed request for nhayesroth October 29, 2025 00:25
@pjannaty

pjannaty commented Oct 29, 2025

Copy link
Copy Markdown
Contributor

Vaibhav tested

#!/bin/bash
export CUDA_VISIBLE_DEVICES="${CUDA_VISIBLE_DEVICES:=0}"
export CHECKPOINT_DIR="${CHECKPOINT_DIR:=./checkpoints}"
export NUM_GPU="${NUM_GPU:=1}"
CUDA_HOME=$CONDA_PREFIX PYTHONPATH=$(pwd) torchrun --nproc_per_node=$NUM_GPU --nnodes=1 --node_rank=0 cosmos_transfer1/diffusion/inference/transfer.py
--checkpoint_dir $CHECKPOINT_DIR
--video_save_folder outputs/single_control_keypoint
--controlnet_specs assets/inference_cosmos_transfer1_single_control_keypoint.json
--offload_text_encoder_model
--offload_guardrail_models
--num_gpus $NUM_GPU

#!/bin/bash
export CUDA_VISIBLE_DEVICES="${CUDA_VISIBLE_DEVICES:=0}"
export CHECKPOINT_DIR="${CHECKPOINT_DIR:=./checkpoints}"
export NUM_GPU="${NUM_GPU:=1}"
CUDA_HOME=$CONDA_PREFIX PYTHONPATH=$(pwd) torchrun --nproc_per_node=$NUM_GPU --nnodes=1 --node_rank=0 cosmos_transfer1/diffusion/inference/transfer.py
--checkpoint_dir $CHECKPOINT_DIR
--video_save_folder outputs/example2_uniform_weights
--controlnet_specs assets/inference_cosmos_transfer1_uniform_weights.json
--offload_text_encoder_model
--offload_guardrail_models
--num_gpus $NUM_GPU

Merging.

@pjannaty
pjannaty merged commit dddf6af into main Oct 29, 2025
4 checks passed
0graph added a commit to 0graph/carla that referenced this pull request Nov 25, 2025
@0graph
fix dockerfile for changes to vllm and transformers (carla-simulator#229
502eb57 
)

nvidia-cosmos/cosmos-transfer1#229
nvidia-cosmos/cosmos-transfer1#210
MarcelPiNacy-CVC pushed a commit to carla-simulator/carla that referenced this pull request Dec 15, 2025
@0graph
Fix issues with running the cosmos transfer server docker image due t…
8ad0bf8 
…o outdate dependencies (#9456)

* fix dockerfile for changes to vllm and transformers (#229)

nvidia-cosmos/cosmos-transfer1#229
nvidia-cosmos/cosmos-transfer1#210

* fix for outdated packages, use new commit hash
berndgassmann added a commit to Motor-Ai/carla-simulator-carla that referenced this pull request Jan 2, 2026
@berndgassmann @MarcelPiNacy-CVC @0graph
Merge upstream ue4-dev (#3)
8ce05e3 
* Extend ROS2 support Step 2: Fine grained ServerSynchronization (carla-simulator#9450)

* Extend ROS2 support Step 2: Fine grained ServerSynchronization

Introduced a fine grained ServerSynchronization mechanism, where each
synchonization participant is treated independently and interacts
with the synchronization of the carla-server individually. If a client
is disconnected (or dies) the synchronization state of all participants
registered via that client are dropped, i.e. the server will continue
running in case the participants of that client were the only ones
demanding synchronous mode.

The synchronization interface provides means of a time window, up to
which the server is allowed to run. Like this, every client can prevent
the carla-server to run too fast depending on their individual speed.
There is no sync-master anymore. Every client decides for its own if
it requires synchronization or not.

Drawback of this change: some existing code might have to be changed
(see removal of synchronous_master in generate_traffic.py).

* Fix Windows build and smoke tests

Update RPC lib version for windows build

Ensure tick calls are ignored if sync mode is not active.
And prevent client changes of fixed_delta_seconds triggering warning
message.

* Client needs to wait for next tick

on non synchronous mode

* Fix libpng (see carlagx's carla-simulator#9469 and berndgassmann's carla-simulator#9451). (carla-simulator#9477)

* Fix issues with running the cosmos transfer server docker image due to outdate dependencies (carla-simulator#9456)

* fix dockerfile for changes to vllm and transformers (carla-simulator#229)

nvidia-cosmos/cosmos-transfer1#229
nvidia-cosmos/cosmos-transfer1#210

* fix for outdated packages, use new commit hash

* Fix docker error due to issues regarding the current working directory. (carla-simulator#9484)

---------

Co-authored-by: MarcelPiNacy-CVC <169088301+MarcelPiNacy-CVC@users.noreply.github.com>
Co-authored-by: 0graph <50416380+0graph@users.noreply.github.com>
donghaiwang pushed a commit to OpenHUTB/hutb that referenced this pull request Jan 26, 2026
donghaiwang
[carla] Fix issues with running the cosmos transfer server docker ima…
570d9aa 
…ge due to outdate dependencies (#9456)

* fix dockerfile for changes to vllm and transformers (#229)

nvidia-cosmos/cosmos-transfer1#229
nvidia-cosmos/cosmos-transfer1#210

* fix for outdated packages, use new commit hash

When running the cosmos-transfer-1 docker container after building there are multiple errors as the gradio server tries to start caused by outdated dependencies.

File "/workspace/cosmos_transfer1/diffusion/module/attention.py", line 24, in <module> from transformer_engine.pytorch.attention.dot_product_attention.dot_product_attention import DotProductAttention ModuleNotFoundError: No module named 'transformer_engine.pytorch.attention.dot_product_attention'; 'transformer_engine.pytorch.attention' is not a package
This commit in the Cosmos-Transfer-1 repository resolves these issues.
nvidia-cosmos/cosmos-transfer1@dddf6af

The Changes I have made:

Updated carla's modified Dockerfile.server to reflect the changed dockerfile in the commit
Updated the commit hash in make_docker.sh as a couple changes in there are also related to the fix.
Where has this been tested?
Platform(s): Ubuntu 22.04 LTS
Python version(s): 3.10
Unreal Engine version(s): 4.26
Had issues over multiple days trying to run my built docker images. after this fix it is working

Possible Drawbacks
users who already have the old repository will not get the new one or the updated dockerfile unless they manually copy it or update the repo.
donghaiwang pushed a commit to OpenHUTB/hutb that referenced this pull request Jan 29, 2026
donghaiwang
[carla] Fix issues with running the cosmos transfer server docker ima…
2a3433c 
…ge due to outdate dependencies (#9456)

* fix dockerfile for changes to vllm and transformers (#229)

nvidia-cosmos/cosmos-transfer1#229
nvidia-cosmos/cosmos-transfer1#210

* fix for outdated packages, use new commit hash

When running the cosmos-transfer-1 docker container after building there are multiple errors as the gradio server tries to start caused by outdated dependencies.

File "/workspace/cosmos_transfer1/diffusion/module/attention.py", line 24, in <module> from transformer_engine.pytorch.attention.dot_product_attention.dot_product_attention import DotProductAttention ModuleNotFoundError: No module named 'transformer_engine.pytorch.attention.dot_product_attention'; 'transformer_engine.pytorch.attention' is not a package
This commit in the Cosmos-Transfer-1 repository resolves these issues.
nvidia-cosmos/cosmos-transfer1@dddf6af

The Changes I have made:

Updated carla's modified Dockerfile.server to reflect the changed dockerfile in the commit
Updated the commit hash in make_docker.sh as a couple changes in there are also related to the fix.
Where has this been tested?
Platform(s): Ubuntu 22.04 LTS
Python version(s): 3.10
Unreal Engine version(s): 4.26
Had issues over multiple days trying to run my built docker images. after this fix it is working

Possible Drawbacks
users who already have the old repository will not get the new one or the updated dockerfile unless they manually copy it or update the repo.
donghaiwang pushed a commit to OpenHUTB/hutb that referenced this pull request Jan 29, 2026
donghaiwang @donghaiwang
[carla] Fix issues with running the cosmos transfer server docker ima…
89e773e 
…ge due to outdate dependencies (#9456)

* fix dockerfile for changes to vllm and transformers (#229)

nvidia-cosmos/cosmos-transfer1#229
nvidia-cosmos/cosmos-transfer1#210

* fix for outdated packages, use new commit hash

When running the cosmos-transfer-1 docker container after building there are multiple errors as the gradio server tries to start caused by outdated dependencies.

File "/workspace/cosmos_transfer1/diffusion/module/attention.py", line 24, in <module> from transformer_engine.pytorch.attention.dot_product_attention.dot_product_attention import DotProductAttention ModuleNotFoundError: No module named 'transformer_engine.pytorch.attention.dot_product_attention'; 'transformer_engine.pytorch.attention' is not a package
This commit in the Cosmos-Transfer-1 repository resolves these issues.
nvidia-cosmos/cosmos-transfer1@dddf6af

The Changes I have made:

Updated carla's modified Dockerfile.server to reflect the changed dockerfile in the commit
Updated the commit hash in make_docker.sh as a couple changes in there are also related to the fix.
Where has this been tested?
Platform(s): Ubuntu 22.04 LTS
Python version(s): 3.10
Unreal Engine version(s): 4.26
Had issues over multiple days trying to run my built docker images. after this fix it is working

Possible Drawbacks
users who already have the old repository will not get the new one or the updated dockerfile unless they manually copy it or update the repo.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@codeJRV codeJRV Awaiting requested review from codeJRV
@spectralflight spectralflight Awaiting requested review from spectralflight

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@rvaibhav @pjannaty