1
0
mirror of https://github.com/tiyn/wiki.git synced 2025-12-12 20:09:45 +01:00

neural networks: added and linked

This commit is contained in:
2025-11-27 08:43:36 +01:00
parent 5d8094e90b
commit 28cf833feb
3 changed files with 38 additions and 2 deletions

View File

@@ -20,8 +20,8 @@ This section addresses the usage of the
#### Preparing TensorFlow Models for the AI HAT+ #### Preparing TensorFlow Models for the AI HAT+
For neural networks to run on the Hailo AI module and the AI HAT+ they have to be converted to the For [neural networks](/wiki/neural_network.md) to run on the Hailo AI module and the AI HAT+ they
`.hef` format. have to be converted to the `.hef` format.
This section assumes the neural network is using This section assumes the neural network is using
[TensorFlow](/wiki/programming_language/python.md#tensorflow) and is available as a `.tf` or [TensorFlow](/wiki/programming_language/python.md#tensorflow) and is available as a `.tf` or
`.tflite` file. `.tflite` file.

32
wiki/neural_network.md Normal file
View File

@@ -0,0 +1,32 @@
# Neural Network
A neural network - often shortened to NN - is a computational model which is part of so called
artificial intelligence.
## Useful Tools
There are some useful tools to use with neural networks.
- [Netron](https://github.com/lutzroeder/netron) is a open-source and self-hostable tool to
visualize neural networks.
It is also easily available from the [official website](https://netron.app/).
## Frameworks
The following is a list of frameworks to design or run neural networks.
- [PyTorch](/wiki/programming_language/python.md#pytorch)
- [Tensorflow](/wiki/programming_language/python.md#tensorflow)
## Neural Network Accelerators
Neural network accelerators are devices to run neural networks by inference.
The following is a list of possible devices.
- [Raspberry Pi AI Hat+](/wiki/linux/raspberry_pi.md#ai-hat) is a standalone accelerator which
directly attaches to a [Raspberry Pi](/wiki/linux/raspberry_pi.md) via the GPIO header.
- [ESP32S3](/wiki/microcontroller.md#esp32) is a self-contained microcontroller by Seeed with
integrated AI acceleration.
It does not need an external connection and is especially useful when a small form-factor is
needed.
- Google Coral Edge TPU is a USB-connected accelerator that is not embedded in the host system.

View File

@@ -162,6 +162,8 @@ This flag is to be used with care.
### PyTorch ### PyTorch
This section addresses the [PyTorch module](https://pytorch.org/). This section addresses the [PyTorch module](https://pytorch.org/).
Pytorch is a machine learning resource which is often used for
[neural networks](/wiki/neural_network.md).
#### Setup Pytorch with Cuda for GPU usage #### Setup Pytorch with Cuda for GPU usage
@@ -190,6 +192,8 @@ This should give back `True`.
### TensorFlow ### TensorFlow
This section addresses the [TensorFlow module](https://www.tensorflow.org/). This section addresses the [TensorFlow module](https://www.tensorflow.org/).
Tensorflos is a machine learning resource which is often used for
[neural networks](/wiki/neural_network.md).
#### Basic Usage of TensorFlow #### Basic Usage of TensorFlow