Machine Learning Lab

User Tools

Site Tools

You are here: Laboratory for Neural Computing for Machine Learning (NCML) » Resources
Trace:
resources

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision		Previous revision
resources [2017/03/30 15:28] – hjresources [2018/06/25 15:32] (current) – hj
Line 4: Line 4:
  
 **Lab Computing Servers**: **Lab Computing Servers**:
 +
 +  * //**data**//:  10-core CPU (i9-7900X) with 128GB memory + 4 GPUs ( 4 x GTX1080 Ti, 11 GB memory each) 
 +
 +  * //**knowledge**//:  10-core CPU (i9-7900X) with 128GB memory + 4 GPUs ( 4 x GTX1080 Ti, 11 GB memory each)
  
   * //**image**//: 6-core CPU (i7-6800K) with 128GB memory + 4 GPUs (4 x GTX 1080, 8GB memory each)   * //**image**//: 6-core CPU (i7-6800K) with 128GB memory + 4 GPUs (4 x GTX 1080, 8GB memory each)
Line 21: Line 25:
   * //**speech1**//: 4-core CPU (i7-6700) with 16G memory    * //**speech1**//: 4-core CPU (i7-6700) with 16G memory 
  
-  * //**speech2**//: 4-core CPU (i7-6700) with 16G memory + 1 GPU (GTX 1070 8GB)+  * //**speech2**//: 4-core CPU (i7-6700) with 16G memory + 1 GPU (GTX 780 Ti 3GB) 
 + 
 +  *  //**speech3**//: 4-core CPU (i7-6700) with 32G memory + 2 GPU  (GTX 1070 8GB + GTX 780 Ti 3GB memory each) 
 + 
 +  *  //**speech4**//: 4-core CPU (i7-6700) with 32G memory + 2 GPUs (GTX 1070 8GB memory each) 
 + 
 +  *  //**speech5**//: 4-core CPU (i7-6700) with 32G memory + 2 GPUs (GTX 1070 8GB memory each) 
 + 
 +  *  //**speech6**//: 4-core CPU (i7-6700) with 32G memory + 2 GPUs (GTX 1070 8GB memory each)
  
-  *  //**speech3**//: 4-core CPU (i7-6700) with 16G memory + 1 GPU  (GTX 780 Ti 3GB memory each)+  *  //**speech7**//: 4-core CPU (i7-6700) with 32G memory + 2 GPUs (2 x GTX 1070 8GB memory each)
  
-  *  //**speech4**//: 4-core CPU (i7-6700) with 16G memory + 2 GPUs (GTX 1070 8GB memory each)+  * //**speech8**//: 4-core CPU (Xeon W-2125) with 32G memory + 256GB SSD + 1 GPU (GTX 1080 8GB memory)
  
-  *  //**speech5**//: 4-core CPU (i7-6700) with 16G memory + 2 GPUs (GTX 1070 8GB memory each)+  * //**speech9**//: 4-core CPU (Xeon W-2125) with 32G memory +  256GB SSD + 1 GPU (GTX 1080 8GB memory)
  
-  *  //**speech6**//: 4-core CPU (i7-6700) with 16G memory + 2 GPUs (GTX 1070 8GB memory each)+  * //**speech10**//: 4-core CPU (Xeon W-2125) with 32G memory + 256GB SSD + 1 GPU (GTX 1080 8GB memory)
  
-  *  //**speech7**//: 4-core CPU (i7-6700) with 16G memory + 2 GPUs (2 x GTX 1070 8GB memory each)+** Check [[:gpu_assignment|the GPU assignments]] before you submit your jobs to GPUs.  
  
-**Check the workload of all CPUs/GPUs before you submit your job(s). [[:check_workload_cpu_gpu|Check here]] for how to check the workloads. Never overload any machines. How to check the workload?** .+[[:check_workload_cpu_gpu|Check here]] for how to check the workloads. Never overload any machines. **
  
 **Storage Space**: **Storage Space**:
Line 52: Line 64:
  
   * **TensorFlow**: see [[:tensorflow|here]] (provided by Mingbin) for how to configure TensureFlow. Alternative, you may configure TensorFlow as [[:tensorflow2|here]] (provided by Chao).   * **TensorFlow**: see [[:tensorflow|here]] (provided by Mingbin) for how to configure TensureFlow. Alternative, you may configure TensorFlow as [[:tensorflow2|here]] (provided by Chao).
 +  * TensorFlow Tutorials:
 +      - [[https://m.youtube.com/watch?feature=youtu.be&v=vq2nnJ4g6N0|Tensorflow and deep learning - without a PhD by Martin Görner]] 
 +      - [[https://m.youtube.com/watch?v=Ejec3ID_h0w&feature=youtu.be|TensorFlow Tutorial (Sherry Moore, Google Brain)]]
 +      - [[http://www.wildml.com/2016/08/rnns-in-tensorflow-a-practical-guide-and-undocumented-features/|RNNs in Tensorflow, a Practical Guide and Undocumented Features – WildML]]
 +      - [[http://warmspringwinds.github.io/tensorflow/tf-slim/2016/12/21/tfrecords-guide/|Tfrecords Guide]]
  
  
 ** * Acknowledgement**: the Tesla graphical card was donated by NVIDIA Inc.  ** * Acknowledgement**: the Tesla graphical card was donated by NVIDIA Inc. 
resources.1490887724.txt.gz · Last modified: 2017/03/30 15:28 by hj