-
Several functions (magnitude, dot, etc) fail because `reduce` has been removed from the core library. To use it, one needs to add:
```python
from functools import reduce
```
to the header
This…
-
I am proposing that a Cross Product function be added to both `Vector2` and `Vector4` of System.Numerics because its use in geometry is so common.
### Rationale and Usage
The structs, `Vector2`, `…
-
In **track.py**, the **class Track** has a **parameter 'feature'** and an **attribute 'features'**. As far as I understood, after a successful association has been done between a detection and a track…
-
### The bug
The Immich backup feature that allows photos to be backed up to the remote server causes the system data on my iPhone 15 Pro Max with iOS 17.6.1 to be filled up completely. This causes th…
-
## ❓ Questions and Help
Hi all,
I am using the pre-trained transformer for my project. I follow ['reproduce ende-wmt14'](https://github.com/facebookresearch/fairseq/issues/346) to train the tran…
-
Hi!
The tutorial is very helpful but when I scale, fit and predict it according to my data set it raises
ValueError: Input contains NaN, infinity or a value too large for dtype('float64').
I hav…
-
- [ ] most parameters are not yet covered.
- [x] vectors (MAD-X array type) are now supported
-
## 🚀 Feature
`F.pairwise_distance` supports `[N,D]x[N,D]=[N]`.
Please add the support for `[*,D]x[*,D]=[*]` with broadcasting.
## Motivation
I am trying to implement a triplet loss.
From th…
-
- we currently use [ForwardDiff](https://github.com/JuliaDiff/ForwardDiff.jl) but should be able to move transparently (= without any user code change) to another backend
- [x] use the backend choice…
-
Instead of maintaining `draw_text` and `draw_label`, it would be nice to just allow the (more capable) draw_label to take vectors of text, x, and y, check the lengths match, and call itself to execute…