train only scannet update

chrischoy · Dec 31, 2019 · bd852b4 · bd852b4
1 parent ee203da
commit bd852b4
Show file tree

Hide file tree

Showing 2 changed files with 6 additions and 5 deletions.
diff --git a/README.md b/README.md
@@ -10,7 +10,7 @@ This repository contains the accompanying code for [4D-SpatioTemporal ConvNets:
 2. Next, preprocess all scannet raw point cloud with the following command after you set the path correctly.
 
 ```
-python -m lib.datasets.prepreocessing.scannet
+python -m lib.datasets.preprocessing.scannet
 ```
 
 3. Train the network with
@@ -65,6 +65,7 @@ The above script trains a network. You have to change the arguments accordingly.
 
 ## Model Zoo
 
-| Model         | Dataset | Voxel Size | Performance              | Link   |
-|:-------------:|:-------:|:----------:|:------------------------:|:------:|
-| Mink16UNet34C | ScanNet | 2cm        | Test set 73.6% mIoU      | [download](https://node1.chrischoy.org/data/publications/minknet/Mink16UNet34C_ScanNet.pth) |
+| Model         | Dataset             | Voxel Size | Conv1 Kernel Size | Performance              | Link   |
+|:-------------:|:-------------------:|:----------:|:-----------------:|:------------------------:|:------:|
+| Mink16UNet34C | ScanNet train + val | 2cm        | 3                 | Test set 73.6% mIoU      | [download](https://node1.chrischoy.org/data/publications/minknet/Mink16UNet34C_ScanNet.pth) |
+| Mink16UNet34C | ScanNet train       | 2cm        | 5                 | Val 72.219% mIoU without rotation average [per class performance](https://github.com/chrischoy/SpatioTemporalSegmentation/issues/13) | [download](https://node1.chrischoy.org/data/publications/minknet/MinkUNet34C-train-conv1-5.pth) |
diff --git a/lib/dataset.py b/lib/dataset.py
@@ -434,7 +434,7 @@ def initialize_data_loader(DatasetClass,
         t.ChromaticAutoContrast(),
         t.ChromaticTranslation(config.data_aug_color_trans_ratio),
         t.ChromaticJitter(config.data_aug_color_jitter_std),
-        t.HueSaturationTranslation(config.data_aug_hue_max, config.data_aug_saturation_max),
+        # t.HueSaturationTranslation(config.data_aug_hue_max, config.data_aug_saturation_max),
     ]
 
   if len(input_transforms) > 0: