图生360度部署 Panodiff

AI视觉网奇2024-12-07 19:35

[1. python build_dataset.py $crop_num$ pairs_num $raw_data_path](#1. python build_dataset.py$ crop_num $pairs_num$ raw_data_path)

[2. Relative Pose Prediction](#2. Relative Pose Prediction)

报错：

[No module named 'pytorch_lightning.utilities.distributed'](#No module named 'pytorch_lightning.utilities.distributed')

[NotImplementedError: No operator found for `memory_efficient_attention_forward` with inputs:](#NotImplementedError: No operator found for memory_efficient_attention_forward with inputs:)

Panodiff

GitHub - shanemankiw/Panodiff: Official implementation for ACM MM 2023 paper '360-Degree Panorama Generation from Few Unregistered NFoV Images'

模型和数据都在提供的压缩包里面。

算法步骤：

1. python build_dataset.py $crop_num$ pairs_num $raw_data_path

2. Relative Pose Prediction

bash 复制代码

GPU=... # Define your GPU Device
save_path=data/pred_results.npy # Define the path for saved predictions

python pred_rotations.py \
    config.yaml \
    --classification_model_path=ckpts/stage1_classification_ckpt.pt \
    --overlap_regression_model_path=ckpts/stage2_overlap_ckpt.pt \
    --nonoverlap_regression_model_path=ckpts/stage2_nonoverlap_ckpt.pt \
    --gpu=$GPU \
    --save_path=$save_path

报错：

No module named 'pytorch_lightning.utilities.distributed'

解决方法：pip install pytorch-lightning==1.5.0

NotImplementedError: No operator found for `memory_efficient_attention_forward` with inputs:

解决方法：

NotImplementedError: No operator found for `memory_efficient_attention_forward` with inputs:-CSDN博客