Update README.md

README.md

@@ -60,7 +60,7 @@ pip install "git+https://github.com/erikwijmans/Pointnet2_PyTorch.git#egg=pointn
 pip install --upgrade https://github.com/unlimblue/KNN_CUDA/releases/download/0.2/KNN_CUDA-0.2-py3-none-any.whl
 ```
 ### Datasets
-For pre-training and fine-tuning, please follow [DATASET.md](https://github.com/lulutang0608/Point-BERT/blob/master/DATASET.md) to install ShapeNet, ModelNet40, ScanObjectNN, and ShapeNetPart datasets, referring to Point-BERT. Speically for Linear SVM evaluation, download the official [ModelNet40](https://shapenet.cs.stanford.edu/media/modelnet40_ply_hdf5_2048.zip) dataset and put the unzip folder under `data/`.
+For pre-training and fine-tuning, please follow [DATASET.md](https://github.com/lulutang0608/Point-BERT/blob/master/DATASET.md) to install ShapeNet, ModelNet40, ScanObjectNN, and ShapeNetPart datasets, referring to Point-BERT. Specially for Linear SVM evaluation, download the official [ModelNet40](https://shapenet.cs.stanford.edu/media/modelnet40_ply_hdf5_2048.zip) dataset and put the unzip folder under `data/`.
 
 The final directory structure should be:
 ```
@@ -84,7 +84,7 @@ Point-M2AE is pre-trained on ShapeNet dataset with the config file `cfgs/pre-tra
 ```bash
 CUDA_VISIBLE_DEVICES=0 python main.py --config cfgs/pre-training/point-m2ae.yaml --exp_name pre-train
 ```
-To evaluate the pre-trained Point-M2AE by Linear SVM on ModelNet40, create the folder `ckpts/` and download the `ckpt-best.pth` into it. Run:
+To evaluate the pre-trained Point-M2AE by Linear SVM on ModelNet40, create the folder `ckpts/` and download the [ckpt-best.pth](https://drive.google.com/file/d/1mkfoGSp01th9Pctlk_mE0o-5sOb3vQpD/view?usp=sharing) into it. You will get 92.87% by running:
 ```bash
 CUDA_VISIBLE_DEVICES=0 python main.py --config cfgs/pre-training/point-m2ae.yaml --exp_name test_svm --test_svm modelnet40 --ckpts ./ckpts/ckpt-best.pth
 ```