-
related paper
|摘要|
|---|
|Very deep convolutional networks have been central to the largest advances in image recognition performance in recent years. One example is the Inception architecture th…
-
**System information**
- What is the top-level directory of the model you are using: Not applicable
- Have I written custom code (as opposed to using a stock example script provided in TensorFlow): …
-
**Test Env:**
Chromium Version: nightly build otcshare/chromium-src@e633308 (75.0.3739.0), otcshare/chromium-src@e7b9a67 (79.0.3917.0)
Platform: macOS 10.14.5
**Expected Result:**
The Image Clas…
-
As people evaluate and experiment with PlaidML I figured it would be useful to start collecting some performance data to help provide some baseline numbers. I'll go first:
### Threadripper 1950X,…
-
This issue is to describe compiler frontend typical positve usage and negative usage.
- positive usage: typical usage of our compiler frontend with supported models.
- negative usage: compiler front…
-
Is it possible to convert the re-trained pytorch model (xx.pth) using the "inception v3" model?
-
**Test Env:**
Chromium Version: nightly build 75.0.3739.0 (**f456c6d**)
Platform: all
**Expected Result:**
The probability shouldn't change.
**Actual Result:**
The probability of the Labels …
-
The InceptionV3 model in PyTorch is trained on ImageNet. According to the official documentation, it uses the mean and standard deviation of ImageNet for normalization. However, I noticed in the code …
-
Like current `luci-value-test` takes long time to test and #12071 is to provide alternative method to reduce time, it would be wonderful if we could reduce test duration somehow.
-
## 시나리오를 다양하게 하여 추론시간 측정
- 각 장비의 최대 처리수를 감안하여 시나리오를 생성 후 측정한다
- 처리할 수 있는 양은 xavier2 > tx2 > nano1 이다. 따라서 3개의 모델에 모두 보낼때는 nano1을 기준으로 하여 시나리오를 생성하고 nano1을 제외하고 보낸다면 tx2를 기준으로 하여 시나리오를 생성한다.
- 각각의…