Hao Shao
Hao Shao
Hi! You just need to specify the paths to the ckpts in this config: https://github.com/opendilab/LMDrive/blob/main/leaderboard/team_code/lmdriver_config.py Maybe this can help you: https://github.com/opendilab/LMDrive/tree/main?tab=readme-ov-file#evaluation
Hi! The Vision encoder pre-training may be proceeded with a 24GB GPU. However, the Instruction fine-tuning stage needs to load a 7B-LLM, which need more GPU memory or more GPU...
Hi! I ran the evaluation on a 3090. I tested it just now and it needed 18567MB GPU memory. I suggest that you run the Carla on the local PC...
Hi! The scenario file is fixed. You can find the route files here: https://github.com/opendilab/LMDrive/tree/main/langauto
Sorry, the current script doesn’t support this feature. But the model (w/o the Carla server) only consume about 14g memory, and 16G v100 is enough. Maybe quantization can be applied...
Hi! Sorry for the late reply, I have been very busy recently. The error you're experiencing seems to originate from the GUI. You can try to comment on the code...
Hi! We will upload all models to openxlab platform in two days!
Hi! @yanfushan We have uploaded the models: https://openxlab.org.cn/models/detail/deepcs233/LMDrive
Hi! Thank you for your attention to this project. However, I have no idea how to provide more details about the evaluation section now. Can you show some problems or...
Hi! You need to install the modified timm package (follow the setup section). ```bash cd vision_encoder pip3 install -r requirements.txt python setup.py develop # if you have installed timm before,...