-
Implementation our defense mechanism "CNN Filter Compression" / "Vector Quantization" (described in the [wiki](https://github.com/Simsso/NIPS-2018-Adversarial-Vision-Challenge/wiki/Defenses-Against-Ad…
-
tensorflow version: 1.8
model: my model structure is very simliar to mobilenet-v1
desc: I trained my model use the tf.contrib.quantize.create_training_graph function, and call the tf.contrib.quantiz…
1icas updated
6 years ago
-
We should investigate the internal features of the tower representation. Doing this should provide several benefits:
- It should help us identify good features that we can compute directly, and fee…
-
I would like to present a few thoughts about the Tensor File Format described in Section 5.2 for consideration.
I expect the rationale for storing a quantization algorithm string in the file is pro…
zoeoz updated
6 years ago
-
the feature request about evolution proposal [OE-14](https://github.com/opencv/opencv/wiki/OE-14.-DNN-FP16)
-
Based on some recent benchmarks (on GTX 1080 Ti), using 8-bit quantized integers during inference could give us significant performance improvement. No experiments has been performed yet, but other so…
-
**TL;DR** There is an edit at the bottom of this post highlighting what is going on because a lot has changed over many discussions, and better alternatives are being explored.
You may want to grab…
-
# Merge
~~embedded_ai/2017-08-07.md at bi-weekly-2017-08-07 · PerfXLab/embedded_ai
https://github.com/PerfXLab/embedded_ai/blob/bi-weekly-2017-08-07/bi-weekly-reports/2017-08-07.md~~
# Comp…
-
Build 'amunmt_compilation_cpu' is failing!
Last 50 lines of build output:
```
[...truncated 6.66 KB...]
/var/lib/jenkins/workspace/amunmt_compilation_cpu/src/amun/3rd_party/fast_align/src/ttables.cc…
-
The network currently runs into out of memory issues at a low number of layers.
This seems to be a problem with TensorFlow's `atrous_conv2d` operation.
If I set the dilation factor to `1`, which means…