add dummy docs

eric-zqwang · Jun 30, 2024 · 9ccdec4 · 9ccdec4
1 parent 1a13b81
commit 9ccdec4
Show file tree

Hide file tree

Showing 5 changed files with 37 additions and 0 deletions.
diff --git a/docs/data_preparation.md b/docs/data_preparation.md
diff --git a/docs/fig/arch.png b/docs/fig/arch.png
diff --git a/docs/installation.md b/docs/installation.md
@@ -0,0 +1,17 @@
+- ### Installation
+**Step 1.** Create conda environment and activate.
+```
+conda create --name puzzlefusionpp python=3.8 -y
+conda activate puzzlefusionpp
+```
+
+**Step 2.** Install PyTorch.
+```
+conda install -y pytorch==2.0.1 torchvision==0.15.2 torchaudio==2.0.2 pytorch-cuda=11.7 -c pytorch -c nvidia
+```
+
+
+**Step 3.** Install PyTorch3d.
+
+
+pip3 install -r requirements.txt
diff --git a/docs/sampling.md b/docs/sampling.md
diff --git a/docs/training.md b/docs/training.md
@@ -0,0 +1,20 @@
+## Training
+
+The training consists of three modules as detailed in the paper. We train the models on 4 Nvidia RTX A6000 GPUs. 
+
+**Stage 1**: VQVAE:
+```
+sh ./scripts/train_vqvae.sh
+```
+
+**Stage 2**: SE3 denoiser:
+```
+sh ./sripts/train_denoiser.sh
+```
+
+**Stage 3**: Pairwise alignment verifier:
+```
+sh ./sripts/train_verifier.sh
+```
+
+We also have provided checkpoint for easier testing [here]().