-
I have 2 ssh-able machines , and this mxnet folder is in the same place on these machines.
When I running:
```
tools/launch.py -n 2 -H hosts python example/image-classification/train_mnist.py --kv-st…
-
Hi,
I have convert tiny imagenet-200 to .rec using im2rec.
I got the following error:
[20:04:27] src/io/iter_image_recordio.cc:212: ImageRecordIOParser: ./train.rec, use 3 threads for decoding..
[20…
-
For I am training on a 11 GB image record file and always get killed because of out of memory.
So I am just wondering that does mxnet load the entire image into the memory?
And what can I do if so.
Th…
-
My system environment:
Mac OS X Yosemite
atlas
no cuDNN
Apple LLVM version 6.1.0 (clang-602.0.49)
Below is the error information:
g++ -c -O3 -Wall -msse2 -Wno-unknown-pragmas -Iinclude -std=c++0x…
-
g++-4.8 -DMSHADOW_FORCE_STREAM -Wall -O3 -I/Users/liguohong01/Code/ml/dl/mxnet/mshadow/ -I/Users/liguohong01/Code/ml/dl/mxnet/dmlc-core/include -fPIC -Iinclude -msse3 -funroll-loops -Wno-unused-parame…
-
Is there a possibility to concatenate mutliple 3 channel images to a set of 9 channel input images while still using dynamic image augmentation (and if possible recordios)?
gz.
-
I can run python train_mnist.py OK. But when i run train_cifar10.py with the following command (I don't have a graphic card) and encountered the error below:
python train_cifar10.py --gpus -1
2015…
-
> > > import mxnet
> > > **\* glibc detected **\* python2.7: free(): invalid pointer: 0x00007fcf34b68880 ***
> > > ======= Backtrace: =========
> > > /lib64/libc.so.6[0x39e4675f4e]
> > > /lib64/libc.s…
-
Any ideas? Installed opencv with macports (homebrew wasn't working) and it is accepted by the makefile but it still complains here and then gives me this NDArray error. I have cuda 7.5 installed on OS…
-
Hi, I am trying to compile proxygen on ubuntu64 with boost installed. I am having a problem with boost_thread. This library is there in /usr/lib however the building system cmake does not find it.
I s…