masahi commented on issue #4497: [WIP] [Relay] Add a PyTorch to Relay Parser
URL: https://github.com/apache/incubator-tvm/pull/4497#issuecomment-582341993
@jwfromm can you have a final look too?
This is an automated message
masahi commented on issue #4497: [WIP] [Relay] Add a PyTorch to Relay Parser
URL: https://github.com/apache/incubator-tvm/pull/4497#issuecomment-582157850
CI is green now, great! Is it ready for final review? @alexwong
This
masahi commented on issue #4497: [WIP] [Relay] Add a PyTorch to Relay Parser
URL: https://github.com/apache/incubator-tvm/pull/4497#issuecomment-580517710
actually I cannot run torchvision tests in this PR on my 8GB laptop. Maybe
RAM is the problem?
masahi commented on issue #4497: [WIP] [Relay] Add a PyTorch to Relay Parser
URL: https://github.com/apache/incubator-tvm/pull/4497#issuecomment-580244271
@alexwong the CI is not updated even if you update the docker script in this
PR (see
masahi commented on issue #4497: [WIP] [Relay] Add a PyTorch to Relay Parser
URL: https://github.com/apache/incubator-tvm/pull/4497#issuecomment-579038361
@alexwong I'm 100% for "make all of the simpler changes to get this merged
first". I don't want you to spend your time redoing the
masahi commented on issue #4497: [WIP] [Relay] Add a PyTorch to Relay Parser
URL: https://github.com/apache/incubator-tvm/pull/4497#issuecomment-578365956
@alexwong I had a big success in refactoring, the parser itself (except op
conversion) is about 150 lines and the "main loop" is just:
masahi commented on issue #4497: [WIP] [Relay] Add a PyTorch to Relay Parser
URL: https://github.com/apache/incubator-tvm/pull/4497#issuecomment-577579947
@alexwong I'm working on loading quantized pytorch models to Relay via QNN.
Your work gave me a great head start :)
masahi commented on issue #4497: [WIP] [Relay] Add a PyTorch to Relay Parser
URL: https://github.com/apache/incubator-tvm/pull/4497#issuecomment-577244252
@alexwong I made a bunch of comments but some of them might not be correct.
I wrote [a modified
masahi commented on issue #4497: [WIP] [Relay] Add a PyTorch to Relay Parser
URL: https://github.com/apache/incubator-tvm/pull/4497#issuecomment-577061566
@alexwong I've sent a PR #4756 which updates PyTorch on CI to 1.4. You can
add ```torch._C._jit_pass_inline(trace.graph)``` after
masahi commented on issue #4497: [WIP] [Relay] Add a PyTorch to Relay Parser
URL: https://github.com/apache/incubator-tvm/pull/4497#issuecomment-574517425
ok I found a dirty hack to remove prim::CallMethod. Add
```torch._C._jit_pass_inline(trace.graph)```
after tracing. This will move
masahi commented on issue #4497: [WIP] [Relay] Add a PyTorch to Relay Parser
URL: https://github.com/apache/incubator-tvm/pull/4497#issuecomment-574403934
I'm happy to merge this even if we can only support v1.3 models for now. I
want to send other op converters I need.
masahi commented on issue #4497: [WIP] [Relay] Add a PyTorch to Relay Parser
URL: https://github.com/apache/incubator-tvm/pull/4497#issuecomment-574076780
@alexwong I tried your PR locally. With pytorch v1.3 it works, but they
introduced big change in
masahi commented on issue #4497: [WIP] [Relay] Add a PyTorch to Relay Parser
URL: https://github.com/apache/incubator-tvm/pull/4497#issuecomment-571958569
@alexwong it seems CI is stuck after failing resnet test?
This is an
13 matches
Mail list logo
