-
`ttnn.permute` fails with inner-most dim = 1 when doing transpose (tested with 2D tensors, e.g. `(2, 1)` with permutation `(1, 0)`).
This is blocking the `aten.t.default` to `ttnn.permute` conversi…
-
Consider the following example:
```julia
xmat = randn(5,5)
xvec = reshape(xmat, 25)
push!(xvec, 3)
xvec[1] = 0.0
xmat[1] # not zero
```
Here `push!` has decoupled `xmat` and `xvec`, though…
-
**Describe the bug**
There is a small issue in deepc.py lines 139 and 140 with the following lines:
u = cp.reshape(u, (self.horizon, self.M))
y = cp.reshape(y, (self.horizon, self.P…
-
模型和转换脚本和 #168 中的相同，toolkit2版本升级到v2.2.0之后又开始报另一个错误：
```log
...
D convert_exmatmul_to_conv: remove node = ['/blocks.23/ffn_key/MatMul'], add node = ['/blocks.23/ffn_key/MatMul']
D convert_…
-
### What happened?
for the given IR
```mlir
module {
func.func @torch_jit(%arg2: !torch.vtensor) -> !torch.vtensor attributes {torch.onnx_meta.ir_version = 7 : si64, torch.onnx_meta.opset_versi…
-
It would be nice to add reshape to wide/long data format like R data.table::melt data.table::dcast or pandas pivot/unpivot and stack/unstack.
Details: https://pandas.pydata.org/docs/user_guide/reshap…
-
Hello, thanks for the nice work,
In the following code, should not the x be permuted first before reshaping,
x = self.ln(x.reshape(b, -1, c))
```
def forward(self, x):
b,c,w,h = x.sha…
-
We are always using `memcpy` for `Reshape` and `ExpandDims` operation.
If these operation's input is used only once and updated shape does not require padding, we can skip copy operation.
For thi…
-
(1) After inner persistent buffers are stored in shared memory. There are still bank conflicts if the persistent buffer is NOT projected to inputs due to two reasons:
```
(a) We are missing a cacheBe…
-
Hi,
When you do this op
You are commiting a mistake.
You are doing `input.view(B, …