Zhang Boheng
Zhang Boheng
How I set weights? If image1 (have class 1 and 2 and class3), the weight is (class1_num + class2_num + class3_num)?
Should I alter dice_coef loss？Is it only for 1-2 classes? Is it better than "categorical_crossentropy loss"? Thanks.
It seems the mask result from mask-rcnn is the same as the rotate rpn of this paper.
I found u say "Submission for Multimodal Brain Tumor Segmentation Challenge 2017 (http://braintumorsegmentation.org/). A patch-based 3D U-Net model is used. Instead of predicting the class label of the center pixel,...
get : E0323 12:40:29.247061 6102 common.cpp:122] Cannot create Cublas handle. Cublas won't be available. E0323 12:40:29.267619 6102 common.cpp:129] Cannot create Curand generator. Curand won't be available. E0323 12:40:29.267684 6102 common.cpp:149]...
hi anlthms, 1.mask = np.array(image > -320, dtype=np.int8) the '-320' is HU value，but u just use it on array value.... 2.coord_z,coord_y,coord_x are the world coordinates.After you run slices = ndimage.interpolation.zoom(data,...
数据集中一个图片有多个caption，但是训练loss计算的时候ground_truth = torch.arange(len(image_logits)).long().to(image_logits.device)，请问不用考虑同图的问题吗？
When I run clip_finetune_flickr.sh during validation, it shows " logits_per_image = (logit_scale * image_features @ text_features.t()).detach().cpu() RuntimeError: CUDA out of memory. Tried to allocate 298.62 GiB (GPU 3; 15.78 GiB...
If I use 4 gpus,: InvalidArgumentError (see above for traceback): Integer division by zero [[Node: training/SGD/gradients/mrcnn_bbox_loss_1/concat_grad/mod = FloorMod[T=DT_INT32, _class=["loc:@mrcnn_bbox_loss_1/concat"], _device="/job:localhost/replica:0/task:0/device:CPU:0"](split_2/split_dim, training/SGD/gradients/mrcnn_bbox_loss_1/concat_grad/Rank)]] [[Node: training/SGD/gradients/tower_3/mask_rcnn/res4t_branch2a/BiasAdd_grad/BiasAddGrad/_13835 = _Recv[client_terminated=false, recv_device="/job:localhost/replica:0/task:0/device:GPU:0", send_device="/job:localhost/replica:0/task:0/device:GPU:3", send_device_incarnation=1, tensor_name="edge_76975_training/SGD/gradients/tower_3/mask_rcnn/res4t_branch2a/BiasAdd_grad/BiasAddGrad", tensor_type=DT_FLOAT,...