[GitHub] [incubator-mxnet] bartekkuncer commented on a change in pull request #20670: [master] Remove MXNET_SUBGRAPH_BACKEND environment variable

[GitBox <gi...@apache.org>]

bartekkuncer commented on a change in pull request #20670:
URL: https://github.com/apache/incubator-mxnet/pull/20670#discussion_r731702543



##########
File path: docs/python_docs/python/tutorials/performance/backend/dnnl/dnnl_readme.md
##########
@@ -233,61 +231,96 @@ Installing the full MKL installation enables MKL support for all operators under
 
   1. Download and install the latest full MKL version following instructions on the [intel website.](https://software.intel.com/en-us/mkl) You can also install MKL through [YUM](https://software.intel.com/content/www/us/en/develop/documentation/installation-guide-for-intel-oneapi-toolkits-linux/top/installation/install-using-package-managers/yum-dnf-zypper.html) or [APT](https://software.intel.com/content/www/us/en/develop/documentation/installation-guide-for-intel-oneapi-toolkits-linux/top/installation/install-using-package-managers/apt.html) Repository.
 
-  2. Run `make -j ${nproc} USE_BLAS=mkl`
+  2. Create and navigate to build directory `mkdir build && cd build`
 
-  3. Navigate into the python directory
+  3. Run `cmake -DUSE_CUDA=OFF -DUSE_BLAS=mkl ..`
 
-  4. Run `sudo python setup.py install`
+  4. Run `make -j`
 
-### Verify whether MKL works
-
-After MXNet is installed, you can verify if MKL BLAS works well with a single dot layer.
+  5. Navigate into the python directory
 
-```
-import mxnet as mx
-import numpy as np
+  6. Run `sudo python setup.py install`
 
-shape_x = (1, 10, 8)
-shape_w = (1, 12, 8)
-
-x_npy = np.random.normal(0, 1, shape_x)
-w_npy = np.random.normal(0, 1, shape_w)
+### Verify whether MKL works
 
-x = mx.sym.Variable('x')
-w = mx.sym.Variable('w')
-y = mx.sym.batch_dot(x, w, transpose_b=True)
-exe = y.simple_bind(mx.cpu(), x=x_npy.shape, w=w_npy.shape)
+After MXNet is installed, you can verify if MKL BLAS works well with a linear matrix solver.
 
-exe.forward(is_train=False)
-o = exe.outputs[0]
-t = o.asnumpy()
+```
+from mxnet import np
+coeff = np.array([[7, 0], [5, 2]])
+y = np.array([14, 18])
+x = np.linalg.solve(coeff, y)
+x.wait_to_read()
 ```
 
 You can open the `MKL_VERBOSE` flag by setting environment variable:

Review comment:
       You can get the verbose log output from mkl library by setting ...




-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: commits-unsubscribe@mxnet.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org