dlbench
dlbench copied to clipboard
hclhkbu
Benchmarking State-of-the-Art Deep Learning Software Tools
Hi! When I use this command : "THC_CACHING_ALLOCATOR=1 CUDA_VISIBLE_DEVICES= 1 THC_CACHING_ALLOCATOR=1 th Main.lua -LR 0.05 -dataset MNIST -network ffn5 -batchSize 342 -epoch 40 -logfile torch_debug.log -save torch_debug.log" The following error...
The architecture of AlexNet implemented in this repo seems to differ greatly from that in the original publication. In your paper, the following is mentioned: "The architecture of AlexNet-R is...
I would like to see how you calculate time in the models. Where can I find this information?
Line 320-324 from file `dlbench/synthetic/experiments/tensorflow/rnn/lstm/` shows that it use real data rather than synthetic data: ``` def main(_): if not FLAGS.data_path: raise ValueError("Must set --data_path to PTB data directory") raw_data...
I can find `t.sh` under: `dlbench/tools/tensorflow/rnn/lstm/t.sh` But I cannot find `tm.sh` or anything seems like a multi-gpu version for LSTM.
Hi Team, I am trying to benchmark a system without gpu. However, while running the benchmark script, it looks for nvidia-smi. ``` CalledProcessError: Command 'nvidia-smi' returned non-zero exit status 127...
I trace here from the paper "Performance Modeling and Evaluation of Distributed Deep Learning Frameworks on GPUs". I am mostly interested into distributed tensorflow's performance, but I only find single...
Hi By reading the paper, I got a bit confused on the concept 'synthetic data'. I noticed that in the paper Fig 2 and Fig 5, both of which are...
From tools/tensorflowbm6_gpu21.config and http://dlbench.comp.hkbu.edu.hk/, we know dlbench of version 8 will call alexnet and resnet for CNN tests. But, Cifar10 experiments won't succeed as the data path is wrong in...
From benchmark.py and configs/*.config, we know dlbench provide capability of changing epoch size if we want a quick test without going through full data set. ``` # From configs/tensorflowbm6_gpu21.config fc;...