Spaces:
Running on L40S
Running on L40S
Download cosmos-framework/examples/launch_sft_vision_super.sh from nvidia/Cosmos3-Action-Viewer: direct link, hf CLI and curl.
- Browser
- Download file 1.85 kB
-
https://huggingface.co/spaces/nvidia/Cosmos3-Action-Viewer/resolve/main/cosmos-framework/examples/launch_sft_vision_super.sh
- Command line
-
hf download hf://spaces/nvidia/Cosmos3-Action-Viewer/cosmos-framework/examples/launch_sft_vision_super.sh
-
curl -L -o launch_sft_vision_super.sh https://huggingface.co/spaces/nvidia/Cosmos3-Action-Viewer/resolve/main/cosmos-framework/examples/launch_sft_vision_super.sh
1.85 kB
| # SPDX-FileCopyrightText: Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved. | |
| # SPDX-License-Identifier: OpenMDW-1.1 | |
| # Structured-TOML launch for vision_sft_super (T2V / I2V / V2V LoRA SFT on | |
| # Qwen3-VL-32B-Instruct, 8-GPU FSDP with CP=2 / DP=4). Drives | |
| # cosmos_framework.scripts.train against examples/toml/sft_config/vision_sft_super.toml. | |
| # | |
| # Optional env vars (defaults below point under examples/; override to put | |
| # data or checkpoints on a different filesystem): | |
| # DATASET_PATH default: examples/data/BridgeData2-Subset-Synthetic-Captions/sft_dataset_bridge | |
| # (must contain train/video_dataset_file.jsonl) | |
| # BASE_CHECKPOINT_PATH default: examples/checkpoints/Cosmos3-Super | |
| # WAN_VAE_PATH default: examples/checkpoints/wan22_vae/Wan2.2_VAE.pth | |
| # HF_TOKEN if any tokenizer download requires gated HF access | |
| # OUTPUT_ROOT default: outputs/train | |
| # | |
| # Usage (8-GPU allocation, inside the training container, from the repo root): | |
| # bash examples/launch_sft_vision_super.sh | |
| TOML_FILE="examples/toml/sft_config/vision_sft_super.toml" | |
| : "${DATASET_PATH:=examples/data/BridgeData2-Subset-Synthetic-Captions/sft_dataset_bridge}" | |
| : "${BASE_CHECKPOINT_PATH:=examples/checkpoints/Cosmos3-Super}" | |
| EXTRA_DATASET_CHECK='[[ -f "$DATASET_PATH/train/video_dataset_file.jsonl" ]] || { echo "ERROR: missing $DATASET_PATH/train/video_dataset_file.jsonl" >&2; exit 1; }' | |
| # Super-variant env tweaks: clear LD_LIBRARY_PATH to avoid host CUDA/NCCL libs | |
| # bleeding into the venv, switch the allocator to expandable_segments so the | |
| # 32B backbone fits without OOM during compile/decode. | |
| export LD_LIBRARY_PATH="" | |
| export PYTORCH_ALLOC_CONF="${PYTORCH_ALLOC_CONF:-expandable_segments:True}" | |
| source "$(dirname "${BASH_SOURCE[0]}")/_sft_launcher_common.sh" | |