josephjaspers / blackcat_tensors

Matrix-Vector Library Designed for Neural Network Construction. cuda (gpu) support, openmp (multithreaded cpu) support, partial support of BLAS, expression template based implementation PTX code generation identical to hand written kernels, and support for auto-differentiation
12 stars 4 forks source link

set_learning_rate crash! #66

Open xinsuinizhuan opened 4 years ago

xinsuinizhuan commented 4 years ago

When i call the set_learning_rate(0.001) to set learning_rate, it crash! 图片

xinsuinizhuan commented 4 years ago

it seems lstm.h set_learning_rate_hook. I don't set learning rate, then to train, when i done, and to save the modle, it also crash: 图片

The thread 0x3518 has exited with code 0 (0x0). The thread 0x4d90 has exited with code 0 (0x0). The thread 0x47ec has exited with code 0 (0x0). Exception thrown at 0x00007FF6A0F2C414 in blackcat_tensors_consoledemo.exe: 0xC00000FD: Stack overflow (parameters: 0x0000000000000001, 0x000000AC366A3FF8). Unhandled exception at 0x00007FF6A0F2C414 in blackcat_tensors_consoledemo.exe: 0xC00000FD: Stack overflow (parameters: 0x0000000000000001, 0x000000AC366A3FF8).

josephjaspers commented 3 years ago

Hello! I am sorry its been a while! I will be looking into these when I can!

I am alive and well. I hope to be able to start working on this again in the near future. I will most likely move the neural networks to a seperate project, but I will maintain both of them!

Sorry for the long delay!

xinsuinizhuan commented 3 years ago

Hello! I am sorry its been a while! I will be looking into these when I can!

I am alive and well. I hope to be able to start working on this again in the near future. I will most likely move the neural networks to a seperate project, but I will maintain both of them!

Sorry for the long delay!

You are well, it is a great good new. I am so worried about you. I am also pay attention to you, you updates your project and star others projection, so i sometimes think you should alive. Protect yourself, stay home when necessary.