-
## Description
I was trying to use TRT modelopt library to quantize a resnet18 from pytorch. The code to reproduce is:
```
from torchvision import models
from torch import nn, optim
# Def…
-
| | |
|--------------------|----|
| Bugzilla Link | [PR36970](https://bugs.llvm.org/show_bug.cgi?id=36970) |
| Status | NEW |
| Importance | P normal |
|…
-
Tobi's performance really goes down as the page size increases. Some of my tests run on very large forms, and 30+ seconds can pass between requests triggered by Tobi.
The issue probably lies somewhe…
-
| | |
|------------------|-----------------|
|Previous ID | SR-4499 |
|Radar | None |
|Original Reporter | @palimondo …
-
Nice solver for beam! Well built and easy to use.
However noticed it is extremely slow for large number of supports. I'm using your code as a benchmark for a FEM solver I'm building to model a beam…
-
I am using Tracy to profile Mutter and GNOME Shell. I would also like to instrument some libraries like GLib or GJS.
These libraries are loaded into many different processes. However, only the part…
-
We are looking at lz4 implementations (and possibly other compressions) for Apache Arrow in https://issues.apache.org/jira/browse/ARROW-8674 and it would be very interesting for us to understand how t…
-
I encountered a weird situation in S2E that with only two alive states but S2E consumed 20GB+ memory, there were some states killed this before, and some symbolic data introduced(say less than 30MB),…
-
Hi, I'm trying to write a very big dataset (millions of entries) to a parquet file. However, there's huge memory allocations happening within this library which leads to server restarts. Using pprof t…
-
**Describe the bug**
When using a dataframe with newer/nullable dtypes, such as the following, I am unable to generate any reports
```python
In[83]: df4.dtypes
Out[83]:
str_col string
…
mands updated
2 years ago