- changed status to resolved
nonftbcnn error while running run_experiments_bcnn_train.m
Hello sir I need your help. Please reply as soon as possible.
After doing settings as mentioned in README for Fine-tuning B-CNN models and running command run_experiments_bcnn_train.m, i am getting this error!
"train: epoch 01: 1/ 24: Error using load Unable to read file 'data/checkgpu/cub-seed-01/nonftbcnn/bcnn_nonft_05705'. No such file or directory."
can u please tell whats the problem?? I dont know what is file "" bcnn_nonft_05705""
I am using Matlab 2017a , cuda 7.5 and ubuntu 14
Full error is :
run_experiments_bcnn_train
ans =
CUDADevice with properties:
Name: 'GeForce GTX 970'
Index: 1
ComputeCapability: '5.2'
SupportsDouble: 1
DriverVersion: 8
ToolkitVersion: 8
MaxThreadsPerBlock: 1024
MaxShmemPerBlock: 49152
MaxThreadBlockSize: [1024 1024 64]
MaxGridSize: [2.1475e+09 65535 65535]
SIMDWidth: 32
TotalMemory: 4.2302e+09
AvailableMemory: 4.0818e+09
MultiprocessorCount: 13
ClockRateKHz: 1253000
ComputeMode: 'Default'
GPUOverlapsTransfers: 1
KernelExecutionTimeout: 0
CanMapHostMemory: 1
DeviceSupported: 1
DeviceSelected: 1
train: epoch 01: 1/ 24: Error using load Unable to read file 'data/checkgpu/cub-seed-01/nonftbcnn/bcnn_nonft_05705'. No such file or directory.
Error in initializeNetworkSharedWeights>getBatch_bcnn_fromdisk (line 201) load(fullfile(imdb.imageDir, imdb.images.name{batch(i)}));
Error in cnn_train>process_epoch (line 293) [im, labels] = getBatch(imdb, batch) ;
Error in cnn_train (line 132) [net,stats.train,prof] = process_epoch(opts, getBatch, epoch, train, learningRate, imdb, net) ;
Error in initializeNetworkSharedWeights (line 161) [netc, info] = cnn_train(netc, bcnndb, @getBatch_bcnn_fromdisk, opts.inittrain, ...
Error in imdb_bcnn_train_dag (line 65) net = initNetFn(imdb, encoderOpts, opts);
Error in run_experiments_bcnn_train (line 79) imdb_bcnn_train_dag(imdb, opts);
Comments (1)
-
reporter - Log in to comment
finally worked :-) Problem was with cuda. Solved magically by some commands. and also had to set useVal = true in model_setup and run_experiments_bcnn_train because i couldn't find out how to make a validation set, its name and path as suggested in readme file "You need to prepare a validation set for the datasets without pre-defined validation set."