jackft2

in make step I meet a error with main.cpp

1

my CmakeList : cmake_minimum_required(VERSION 3.0 FATAL_ERROR) project(yolo-app) find_package( OpenCV 3.4.9 REQUIRED ) include_directories( ${OpenCV_INCLUDE_DIRS} ) message(${OpenCV_INCLUDE_DIRS}) find_package(Torch REQUIRED) aux_source_directory(. DIR_SRCS) add_executable(yolo-app ${DIR_SRCS}) target_link_libraries(yolo-app "${TORCH_LIBRARIES}" "${OpenCV_LIBS}") set_property(TARGET yolo-app PROPERTY CXX_STANDARD 11)...

如題

> > > > I met the same error, did you solve it ? > > > > > > > > > Yes I have solved it > >...

hellow I prepare my own data according your suggestion , but it's fault.

8

Traceback (most recent call last): File "train.py", line 367, in main() File "train.py", line 327, in main train_log = train(config, train_loader, model, criterion, optimizer) File "train.py", line 112, in train...

as title

i met a problem when i enter python3 setup.py build develop

3

/usr/local/lib/python3.6/dist-packages/torch/include/ATen/core/TensorBody.h:354:1: note: declared here T * data() const { ^ ~~ /home/ubuntu/workspace/MaskTextSpotterV3/maskrcnn_benchmark/csrc/cuda/SigmoidFocalLoss_cuda.cu:173:880: warning: ‘T* at::Tensor::data() const [with T = float]’ is deprecated: Tensor.data() is deprecated. Please use Tensor.data_ptr() instead. [-Wdeprecated-declarations]...

can we convert the model by torchscript?

1

above all

what's different between predictions and seg_predictions? what's different between words and seq_words?

do you recommend any label tool for training MaskTextSpotterV3

1

above all

ModuleNotFoundError: No module named 'mvtec'

1

what's package or anything others did I need to install