-
Notifications
You must be signed in to change notification settings - Fork 3.5k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[FRONTEND][TENSORFLOW] Bugfix #2326
Conversation
@@ -646,6 +646,9 @@ def _transform_mask(stride_dim, ellipsis_mask): | |||
pass | |||
else: | |||
final_output.append(out_shape[gather_index]) | |||
# Prevent 0-dim tensors which are not accepted by nnvm |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@dominicsymes thanks for the change.
Please add a test case for this.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@srkreddy1238 thanks for taking a look.
I have added a test case that triggers strided slice with 0-dim output.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks @dominicsymes . LGTM.
commit f347b52 Author: Yong Wu <yongwu@amazon.com> Get tags of saved model automatically commit 916576c Author: Zhi Chen <chzhi@amazon.com> Support TensorFlow saved model TF parser: return the consistent error message to error handler commit f1782f3 Author: Yong Wu <yongwu@amazon.com> Add tf parser wrapper, infer shape automatically commit 76188a4 Author: Siva <sivar.b@huawei.com> [NNVM][TENSORFLOW] bugfix. (apache#2444) commit 6737739 Author: Ashutosh Parkhi <ashutosh.parkhi@imgtec.com> [Tensorflow] Support for Crop (apache#2285) commit f6c3f99 Author: Alexey Romanov <alexey.v.romanov@gmail.com> [FRONTEND][TENSORFLOW] Use input shapes directly instead of 1-element lists (apache#2242) commit e5d92e1 Author: Dominic Symes <36929632+dominicsymes@users.noreply.github.com> [FRONTEND][TENSORFLOW] Bugfix (apache#2326) commit 00d509d Author: Alexey Romanov <alexey.v.romanov@gmail.com> [FRONTEND][TENSORFLOW] Support Unstack and Split (apache#2105) commit df9d3ad Author: Siva <sivar.b@huawei.com> [FRONTEND][TENSORFLOW] Bugfix (apache#2267) commit d1a0c90 Author: Zhebin Jin <zhebin.jzb@alibaba-inc.com> [FRONTEND][TENSORFLOW]Add Split and realdiv op support (apache#2123) * Add Split and realdiv op support * Fix the pad calculation in the case of dilated convolution
commit f347b52 Author: Yong Wu <yongwu@amazon.com> Get tags of saved model automatically commit 916576c Author: Zhi Chen <chzhi@amazon.com> Support TensorFlow saved model TF parser: return the consistent error message to error handler commit f1782f3 Author: Yong Wu <yongwu@amazon.com> Add tf parser wrapper, infer shape automatically commit 76188a4 Author: Siva <sivar.b@huawei.com> [NNVM][TENSORFLOW] bugfix. (apache#2444) commit 6737739 Author: Ashutosh Parkhi <ashutosh.parkhi@imgtec.com> [Tensorflow] Support for Crop (apache#2285) commit f6c3f99 Author: Alexey Romanov <alexey.v.romanov@gmail.com> [FRONTEND][TENSORFLOW] Use input shapes directly instead of 1-element lists (apache#2242) commit e5d92e1 Author: Dominic Symes <36929632+dominicsymes@users.noreply.github.com> [FRONTEND][TENSORFLOW] Bugfix (apache#2326) commit 00d509d Author: Alexey Romanov <alexey.v.romanov@gmail.com> [FRONTEND][TENSORFLOW] Support Unstack and Split (apache#2105) commit df9d3ad Author: Siva <sivar.b@huawei.com> [FRONTEND][TENSORFLOW] Bugfix (apache#2267) commit d1a0c90 Author: Zhebin Jin <zhebin.jzb@alibaba-inc.com> [FRONTEND][TENSORFLOW]Add Split and realdiv op support (apache#2123) * Add Split and realdiv op support * Fix the pad calculation in the case of dilated convolution
commit f347b52 Author: Yong Wu <yongwu@amazon.com> Get tags of saved model automatically commit 916576c Author: Zhi Chen <chzhi@amazon.com> Support TensorFlow saved model TF parser: return the consistent error message to error handler commit f1782f3 Author: Yong Wu <yongwu@amazon.com> Add tf parser wrapper, infer shape automatically commit 76188a4 Author: Siva <sivar.b@huawei.com> [NNVM][TENSORFLOW] bugfix. (apache#2444) commit 6737739 Author: Ashutosh Parkhi <ashutosh.parkhi@imgtec.com> [Tensorflow] Support for Crop (apache#2285) commit f6c3f99 Author: Alexey Romanov <alexey.v.romanov@gmail.com> [FRONTEND][TENSORFLOW] Use input shapes directly instead of 1-element lists (apache#2242) commit e5d92e1 Author: Dominic Symes <36929632+dominicsymes@users.noreply.github.com> [FRONTEND][TENSORFLOW] Bugfix (apache#2326) commit 00d509d Author: Alexey Romanov <alexey.v.romanov@gmail.com> [FRONTEND][TENSORFLOW] Support Unstack and Split (apache#2105) commit df9d3ad Author: Siva <sivar.b@huawei.com> [FRONTEND][TENSORFLOW] Bugfix (apache#2267) commit d1a0c90 Author: Zhebin Jin <zhebin.jzb@alibaba-inc.com> [FRONTEND][TENSORFLOW]Add Split and realdiv op support (apache#2123) * Add Split and realdiv op support * Fix the pad calculation in the case of dilated convolution
commit 76188a4 Author: Siva sivar.b@huawei.com [NNVM][TENSORFLOW] bugfix. (apache#2444) commit 6737739 Author: Ashutosh Parkhi ashutosh.parkhi@imgtec.com [Tensorflow] Support for Crop (apache#2285) commit f6c3f99 Author: Alexey Romanov alexey.v.romanov@gmail.com [FRONTEND][TENSORFLOW] Use input shapes directly instead of 1-element lists (apache#2242) commit e5d92e1 Author: Dominic Symes 36929632+dominicsymes@users.noreply.github.com [FRONTEND][TENSORFLOW] Bugfix (apache#2326) commit 00d509d Author: Alexey Romanov alexey.v.romanov@gmail.com [FRONTEND][TENSORFLOW] Support Unstack and Split (apache#2105) commit df9d3ad Author: Siva sivar.b@huawei.com [FRONTEND][TENSORFLOW] Bugfix (apache#2267) commit d1a0c90 Author: Zhebin Jin zhebin.jzb@alibaba-inc.com [FRONTEND][TENSORFLOW]Add Split and realdiv op support (apache#2123) * Add Split and realdiv op support * Fix the pad calculation in the case of dilated convolution
commit 76188a4 Author: Siva sivar.b@huawei.com [NNVM][TENSORFLOW] bugfix. (apache#2444) commit 6737739 Author: Ashutosh Parkhi ashutosh.parkhi@imgtec.com [Tensorflow] Support for Crop (apache#2285) commit f6c3f99 Author: Alexey Romanov alexey.v.romanov@gmail.com [FRONTEND][TENSORFLOW] Use input shapes directly instead of 1-element lists (apache#2242) commit e5d92e1 Author: Dominic Symes 36929632+dominicsymes@users.noreply.github.com [FRONTEND][TENSORFLOW] Bugfix (apache#2326) commit 00d509d Author: Alexey Romanov alexey.v.romanov@gmail.com [FRONTEND][TENSORFLOW] Support Unstack and Split (apache#2105) commit df9d3ad Author: Siva sivar.b@huawei.com [FRONTEND][TENSORFLOW] Bugfix (apache#2267) commit d1a0c90 Author: Zhebin Jin zhebin.jzb@alibaba-inc.com [FRONTEND][TENSORFLOW]Add Split and realdiv op support (apache#2123) * Add Split and realdiv op support * Fix the pad calculation in the case of dilated convolution
* [FRONTEND][TENSORFLOW] Enhance with left over patches from NNVM. commit 76188a4 Author: Siva sivar.b@huawei.com [NNVM][TENSORFLOW] bugfix. (#2444) commit 6737739 Author: Ashutosh Parkhi ashutosh.parkhi@imgtec.com [Tensorflow] Support for Crop (#2285) commit f6c3f99 Author: Alexey Romanov alexey.v.romanov@gmail.com [FRONTEND][TENSORFLOW] Use input shapes directly instead of 1-element lists (#2242) commit e5d92e1 Author: Dominic Symes 36929632+dominicsymes@users.noreply.github.com [FRONTEND][TENSORFLOW] Bugfix (#2326) commit 00d509d Author: Alexey Romanov alexey.v.romanov@gmail.com [FRONTEND][TENSORFLOW] Support Unstack and Split (#2105) commit df9d3ad Author: Siva sivar.b@huawei.com [FRONTEND][TENSORFLOW] Bugfix (#2267) commit d1a0c90 Author: Zhebin Jin zhebin.jzb@alibaba-inc.com [FRONTEND][TENSORFLOW]Add Split and realdiv op support (#2123) * Add Split and realdiv op support * Fix the pad calculation in the case of dilated convolution * * review comments * * resnet fix. * * review comments
…che#2757) * [FRONTEND][TENSORFLOW] Enhance with left over patches from NNVM. commit 76188a4 Author: Siva sivar.b@huawei.com [NNVM][TENSORFLOW] bugfix. (apache#2444) commit 6737739 Author: Ashutosh Parkhi ashutosh.parkhi@imgtec.com [Tensorflow] Support for Crop (apache#2285) commit f6c3f99 Author: Alexey Romanov alexey.v.romanov@gmail.com [FRONTEND][TENSORFLOW] Use input shapes directly instead of 1-element lists (apache#2242) commit e5d92e1 Author: Dominic Symes 36929632+dominicsymes@users.noreply.github.com [FRONTEND][TENSORFLOW] Bugfix (apache#2326) commit 00d509d Author: Alexey Romanov alexey.v.romanov@gmail.com [FRONTEND][TENSORFLOW] Support Unstack and Split (apache#2105) commit df9d3ad Author: Siva sivar.b@huawei.com [FRONTEND][TENSORFLOW] Bugfix (apache#2267) commit d1a0c90 Author: Zhebin Jin zhebin.jzb@alibaba-inc.com [FRONTEND][TENSORFLOW]Add Split and realdiv op support (apache#2123) * Add Split and realdiv op support * Fix the pad calculation in the case of dilated convolution * * review comments * * resnet fix. * * review comments
…che#2757) * [FRONTEND][TENSORFLOW] Enhance with left over patches from NNVM. commit 76188a4 Author: Siva sivar.b@huawei.com [NNVM][TENSORFLOW] bugfix. (apache#2444) commit 6737739 Author: Ashutosh Parkhi ashutosh.parkhi@imgtec.com [Tensorflow] Support for Crop (apache#2285) commit f6c3f99 Author: Alexey Romanov alexey.v.romanov@gmail.com [FRONTEND][TENSORFLOW] Use input shapes directly instead of 1-element lists (apache#2242) commit e5d92e1 Author: Dominic Symes 36929632+dominicsymes@users.noreply.github.com [FRONTEND][TENSORFLOW] Bugfix (apache#2326) commit 00d509d Author: Alexey Romanov alexey.v.romanov@gmail.com [FRONTEND][TENSORFLOW] Support Unstack and Split (apache#2105) commit df9d3ad Author: Siva sivar.b@huawei.com [FRONTEND][TENSORFLOW] Bugfix (apache#2267) commit d1a0c90 Author: Zhebin Jin zhebin.jzb@alibaba-inc.com [FRONTEND][TENSORFLOW]Add Split and realdiv op support (apache#2123) * Add Split and realdiv op support * Fix the pad calculation in the case of dilated convolution * * review comments * * resnet fix. * * review comments
Hello. I noticed a couple of problems while compiling an example with the tensorflow frontend module. The two small changes in the attached commit seem to fix the issues. One change is a simple rename of a dummy variable which was clashing with the function argument 'shape'. The second is to deal with the case of when a shrink axis operator produces a tensor with dimension zero - this does not seem to work with nnvm where the dimension must be at least one. I'd welcome any comments - particularly on whether this is the correct way to deal with the 0-dim result. Thanks.