Skip to content

Commit

Permalink
[WebNN EP] Disable ConvTranspose for WebNN CPU (#20762)
Browse files Browse the repository at this point in the history
WebNN CPU backend implementation has been migrated from XNNPack to
TFLite, currently TFLite has not supported WebNN's convTranspose2d yet,
just disable it for now.
  • Loading branch information
Honry authored May 23, 2024
1 parent 529feb0 commit 2c39d0c
Show file tree
Hide file tree
Showing 2 changed files with 2 additions and 2 deletions.
2 changes: 1 addition & 1 deletion js/web/docs/webnn-operators.md
Original file line number Diff line number Diff line change
Expand Up @@ -22,7 +22,7 @@ operators and the supported opset domain/versions in **WebNN EP** by ONNX Runtim
| Clip | ai.onnx(7-10, 11, 12, 13+) | clamp ||| |
| Concat | ai.onnx(7-10, 11-12, 13+) | concat ||| |
| Conv | ai.onnx(7-10, 11+) | conv2d ||| Only supports 3-D or 4-D input and 'W' (weight). WebNN CPU requires the 'W' (weight) input to be a constant |
| ConvTranspose | ai.onnx(7-10, 11+) | convTranspose2d || | Only supports 3-D or 4-D input and 'W' (weight). WebNN CPU requires the 'W' (weight) input to be a constant |
| ConvTranspose | ai.onnx(7-10, 11+) | convTranspose2d || | Only supports 3-D or 4-D input and 'W' (weight). |
| Cos | ai.onnx(7+) | cos ||| |
| Div | ai.onnx(7-12, 13, 14+) | div ||| |
| Elu | ai.onnx(7+) | elu ||| |
Expand Down
2 changes: 1 addition & 1 deletion onnxruntime/core/providers/webnn/builders/helper.h
Original file line number Diff line number Diff line change
Expand Up @@ -167,7 +167,7 @@ static const InlinedHashMap<std::string, WebnnOpInfo> op_map = {
{"Concat", {"concat", true}},
{"Conv", {"conv2d", true}},
{"ConvInteger", {"conv2dInteger", false}},
{"ConvTranspose", {"convTranspose2d", true}},
{"ConvTranspose", {"convTranspose2d", false}},
{"Cos", {"cos", false}},
{"Div", {"div", true}},
{"DequantizeLinear", {"dequantizeLinear", false}},
Expand Down

0 comments on commit 2c39d0c

Please sign in to comment.