-
Get the SVN and later the neural network to run on the ESP32 and test performance.
Requires developing a process to compress the results of machine learning and integrating it into the firmware.
Res…
xjjak updated
3 months ago
-
# KD History
## History
**Pattern**
```
- 2020-05 **[Name]** paper_full_name [[Paper](paperlink)] [[Code](codelink)] [[Note](notelink)] #tags
```
- 2006 Model Compression [[Paper](http://www.c…
-
- https://arxiv.org/abs/1911.02497
- 2019
深層ニューラルネットワーク(DNN)は、学習したタスクに必要な量よりもはるかに多くの重みを高精度で表現していることが多い。
そのため、重みの刈り込みや量子化などの手法を用いて圧縮することで、精度を落とさずにモデルサイズと推論時間を削減できることが多い。
しかし、DNN、ハードウェアプラットフォーム、最…
e4exp updated
2 years ago
-
-
Hi, thank you for creating this repo. It is super helpful for following the research in this field!
I think there are several related papers that are missing from the repo:
**Survey-&-Theory**:
-…
-
## 一言でいうと
機械学習モデルをより低コストで動かすための研究についてのサーベイ。冗長パラメーターの削除、行列因子分解による計算コスト削減、省エネ特化モデルの作成、蒸留といった4つの類型に分類されて紹介されている。
![image](https://user-images.githubusercontent.com/544269/34589658-37246f28-f1f5-11e…
-
[cxx/ClassResultList.cxx](https://github.com/SwuduSusuwu/SubStack/blob/trunk/cxx/ClassResultList.cxx) has correspondances to [neocortex](https://wikipedia.org/wiki/Neocortex). which is what humans use…
-
Hello,
I'm attempting to train a model for a micro-controller that only supports 8-bit precision or lower.
This works perfectly when training using your `QuantizationAwareTrainingConfig`.
In add…
-
Hi there, 1st of all thanks for this awesome work !
Since we've 'doxed' it in our [HyMPS project](https://forart.it/HyMPS) (under [VIDEO section](https://github.com/FORARTfe/HyMPS#-1 ) \ [AI-based …
-
# General info
- [x] Conference: ICPR2020 + RRPR2020
- [x] Submission date: July 2020 (ICPR) and Nov 20th, 2020 (RRPR)
- [x] Papers title: "Compression strategies and space-conscious representation…