cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
yang.liu@avnet.com
Participant
Participant
871 Views
Registered: ‎10-23-2017

Does anyone have the same error message as me?

Jump to solution

I follow the steps of https://github.com/Xilinx/ml-suite/blob/v1.4/examples/caffe/README.md .

try this command:

python run.py --prototxt /opt/models/caffe/bvlc_googlenet/bvlc_googlenet_train_val.prototxt --caffemodel /opt/models/caffe/bvlc_googlenet/bvlc_googlenet.caffemodel --prepare --output_dir work

error message appears:

**************************************************
* BUILDING DATA FLOW GRAPH
**************************************************
Traceback (most recent call last):
File "run.py", line 131, in <module>
Compile(args["output_dir"])
File "run.py", line 59, in Compile
compiler.compile()
File "/scratch/MLsuite/xfdnn/tools/compile/bin/xfdnn_compiler_caffe.py", line 198, in compile
File "/scratch/MLsuite/xfdnn/tools/compile/optimizations/changeinplace.py", line 42, in read_file
IOError: [Errno 2] No such file or directory: 'work/deploy.prototxt'

 

Why do I get this error when I follow the steps completely? 

Please help me on this.  Thank you!

Tags (1)
0 Kudos
1 Solution

Accepted Solutions
yang.liu@avnet.com
Participant
Participant
774 Views
Registered: ‎10-23-2017

I tried to use a new server which support AVX2.0.  It works perfectly.

My previous server is a bit old that can not support AVX2.0. 

The quantization step uses AVX2.0.  This error may be caused by the CPU which can not support the AVX2.0 instruction set. 

View solution in original post

3 Replies
xilinxacct
Instructor
Instructor
845 Views
Registered: ‎10-23-2018

yang.liu@avnet.com 

Does the directory 'work' already exist and in the relative position for the project? If not, try creating the directory, then retry the operation.

Hope that Helps
If so, Please mark as solution accepted. Kudos also welcomed. :-)

yang.liu@avnet.com
Participant
Participant
810 Views
Registered: ‎10-23-2017

Hi xilinxacct,

Thanks for your reply.

But the work directory was already created by Xilinx. It's a Inherent folder in docker image.

There's just a Readme file in work directory. So I guess deploy.prototxt was not successfully generated by previous steps. 

So any suggestions? 

Thank you!

0 Kudos
yang.liu@avnet.com
Participant
Participant
775 Views
Registered: ‎10-23-2017

I tried to use a new server which support AVX2.0.  It works perfectly.

My previous server is a bit old that can not support AVX2.0. 

The quantization step uses AVX2.0.  This error may be caused by the CPU which can not support the AVX2.0 instruction set. 

View solution in original post