guberti commented on code in PR #13242:
URL: https://github.com/apache/tvm/pull/13242#discussion_r1040177225


##########
python/tvm/topi/arm_cpu/mprofile/dsp/micro_kernel/tensordot.py:
##########
@@ -14,142 +14,416 @@
 # KIND, either express or implied.  See the License for the
 # specific language governing permissions and limitations
 # under the License.
-"""Computes a "jumpy tensordot" operator, which can be used to tensorize many 
common operators
-including regular conv2d, depthwise conv2d, and grouped conv2d provided the 
data and kernel layouts
-are the optimal ones. When groups=1, the optimal data layout is NHWC and 
kernel layout is OHWI. When
-this is a depthwise convolution, the optimal data layout is NCHW and kernel 
layout is OIHW."""
+"""Generates optimized code to compute a tensor dot product on ARMv7E-M.

Review Comment:
   Sometimes - this uses the DSP instructions, which are required in v7E-M but 
optional in v8-M. This code also does not use MVE, which is optional in v8-M 
but would be really useful for deep learning.



-- 
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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to