liheyuan
I suggest both training loss function without KD and with KD should add a softmax function, because the outputs of models are without softmax. Just like this. https://github.com/peterliht/knowledge-distillation-pytorch/blob/e4c40132fed5a45e39a6ef7a77b15e5d389186f8/model/net.py#L100-L114 ==> `KD_loss...
I only generate car only data when I prepare the data. Then I use the car only data to train and test the v2 model(using default hyper-parameters). However, I cannot...
https://github.com/charlesq34/frustum-pointnets/blob/2ffdd345e1fce4775ecb508d207e0ad465bcca80/train/provider.py#L224-L227 I don't know why adding np.pi/2.0 can help us to get the center view? Why does shifting the rot angle by pi/2 can be helpful to adjust GT heading...
Dear author, @anhttran For unknown reasons, my 3d reconstructed faces don't have fine details. this is an image shown in your paper:  this is what I get by run...
Thank you so much for sharing your amazing code! Will you release the code of UV-texture GAN mentioned in the Supplemental Materials of your paper?
When I tried to run the example ``` python train.py --dataset=ucf101 --model=resnet --video_path=/home/tomrunia/data/UCF-101/jpg --annotation_path=/home/tomrunia/data/UCF-101/ucfTrainTestlist/ucf101_01.json --batch_size=64 --num_classes=101 --momentum=0.9 --weight_decay=1e-3 --model_depth=34 --resnet_shortcut=A --spatial_size=112 --sample_duration=16 --optimizer=SGD --learning_rate=0.01 ``` there is an error: ```...
https://github.com/Halfish/cs231n/blob/d355c0c61296d80c7709907271c548d2994e9990/assignment2/cs231n/layers.py#L506-L507 https://github.com/Halfish/cs231n/blob/d355c0c61296d80c7709907271c548d2994e9990/assignment2/cs231n/layers.py#L540-L541 应为： ``` out_height = 1 + (H - pool_height) / stride out_width = 1 + (W - pool_width) / stride ```
I notice that in compute_shape(), you only use mean shape, but not mean expression. I understand that it is because you use an expression basis provided by Guo et al,...