-
-
For the example below, comparing GPU with CPU speed shows that GPU is very slow.
https://github.com/oneapi-src/oneDNN/blob/master/examples/primitives/batch_normalization.cpp
```
std::chrono::system…
-
# Overview
v0.7 is brings many major features. The community works together to refactor the internal code base to bring an unified IR code structure with unified IRModule, type system and pass infr…
-
## Description
Any keras model exported to onnx using group normalization crashes tensorrt.
## Environment
**TensorRT Version**: 7
**GPU Type**: 2080TI
**Nvidia Driver Version**: 440.33
…
-
I used Larq to create a binary resnet20 and a full precision resnet20 by setting the input quantiser and kernel quantiser to 'set_sign' and None respectively. I then converted the .h5 file into tflite…
-
Stock Ubuntu 19.04 with Cuda 10.0, Tensorflow 2.0.0 installed via pip3, Python 3.7.3, GTX1060.
I have a float64 valued dataset with a simple conv2d network that includes tf.keras.layers.BatchNormal…
tb438 updated
4 years ago
-
## ❓ Questions and Help
During reduce, it seems DGL makes multiple calls to the provided reduce function during .recv(), where in each call a group of nodes with mailboxes of the same size is proce…
-
@shubha-ramani @SeverineH @alalek
**config**:ubuntu16.04 python3 tf1.12 opnvino R5.1
**the repo**:https://github.com/LuWei6896/text-detection-ctpn
**the model file**:https://drive.google.com/file/d…
-
**Describe the bug**
Building from the Dockerfile.cuda and Dockerfile.source files, get the following error failure:
```
Processing /root/onnxruntime_gpu-1.7.0-cp36-cp36m-linux_x86_64.whl
Collecti…
-
Please make sure that this is a bug. As per our
[GitHub Policy](https://github.com/tensorflow/tensorflow/blob/master/ISSUES.md),
we only address code/doc bugs, performance issues, feature requests a…