-
# 🌟 New SSL approach addition
## Approach description
NNCLR: https://arxiv.org/abs/2104.14548
> Self-supervised learning algorithms based on instance discrimination train encoders to be invar…
-
Hi, except NNCLR, other models faced this error `ValueError: operands could not be broadcast together with shapes (32,) (6,) (32,) `
Here is the log:
```
/usr/local/lib/python3.6/dist-packages/…
-
We are in the process of updating our benchmark code and adding benchmarks on the ImageNet dataset. The goal of this issue is to coordinate and track progress of the update.
The new benchmark code …
-
Some models in **lightly** used **SimCLRTransform** (e.g, DCL, NNCLR); however, these models also employ the idea of multi-crop (that's originally used for SwaV model) to improve their results, as men…
-
The main goal is to stabilize KerasCV's integration with Keras 3. The list of outstanding items listed below needs to be adresses to make KerasCV feature complete with Keras 3 is listed below. Please …
-
**Problem**
When I try to run CIFAR datasets with Swin-Tiny Backbone, it return the error:
"AssertionError: Input image height (32) doesn't match model (224)."
**Reproducibility**
I'm using the…
-
-
- https://arxiv.org/abs/2104.14548
- 2021
インスタンス判別に基づく自己教師付き学習アルゴリズムは、同じインスタンスの事前定義された変換に対して不変であるようにエンコーダを訓練します。
多くの手法では、同じ画像の異なるビューをコントラスト損失のためのポジティブとして扱いますが、我々はデータセット内の他のインスタンスからのポジティブを使用することに興…
e4exp updated
3 years ago
-
# Type check lightly.models
* Work through the excluded files with prefix lightly.models and tests/models in [mypy.ini](https://github.com/lightly-ai/lightly/blob/master/mypy.ini) and type check one …
-
We need to convert keras.io examples to work with Keras 3.
This involves two stages:
## Stage 1: tf.keras backwards compatibility check
Keras 3 is intended as a drop-in replacement for tf.ker…