-
For queries like `select * from a, b on a.x = b.x and a.y = b.x`, we can push down a filter before side `a` with condition `a.x=a.y` and only keep one of the two join condition in hash join executor.
…
-
png by default is a reasonable choice. However, it's surely doesn't offer the best compression as of today. For instance, lossless webp consistently beats png in terms of compression ratio.
I sugge…
-
I'm still not quite understanding why it's acceptable to have TLA silently change the evaluation order.
If you `import defer` a module graph, subgraphs that contain TLA are eagerly evaluated. So, w…
-
Code coverage reports from the VM are based on invocation counters on functions (Function) and call sites (ICData). If the invocation counter is non-zero, the position is a hit, otherwise a miss. When…
-
Hi,
When compiling my model, I get `ValueError(graph contains cycles)` during the check in `general_toposort`. The traceback is long, but the beginning is:
```
ERROR (theano.gof.opt): SeqOptimizer …
-
Hello,
Maybe I am not looking in the right place, but is there a way to query the confidence value of a given robot position measurement in the pose graph?
Thank you for your answer!
-
The followings bugs appeared in our first installation on Tomcat:
1. white background
2. graphs missing
3. settings and tools options doesn't work
4. at the top like the options name "Servers" "Alerts…
-
### 🐛 Describe the bug
Hi team,
we're now investigating the export to onnx feature and we found that some update logic in the original pytorch model is not working in the converted onnx model. The …
-
### Describe the issue
ONNX opset version: 14
When adding an extra **torch.Tensor.transpose** node as output, the original output of **torch.Tensor.to** is changed with and without optimization,…
Azyka updated
11 months ago
-
Nuitka has too little in the way of measuring the actual performance gains one has. You would change that.
In a first stage, you would enhance the existing http://speedcenter.nuitka.net to provide …