This is the Linux app named Model Search whose latest release can be downloaded as model_searchsourcecode.tar.gz. It can be run online in the free hosting provider OnWorks for workstations.
Download and run online this app named Model Search with OnWorks for free.
Befolgen Sie diese Anweisungen, um diese App auszuführen:
- 1. Diese Anwendung auf Ihren PC heruntergeladen.
- 2. Geben Sie in unserem Dateimanager https://www.onworks.net/myfiles.php?username=XXXXX den gewünschten Benutzernamen ein.
- 3. Laden Sie diese Anwendung in einem solchen Dateimanager hoch.
- 4. Starten Sie den OnWorks Linux-Online- oder Windows-Online-Emulator oder den MACOS-Online-Emulator von dieser Website.
- 5. Rufen Sie vom gerade gestarteten OnWorks Linux-Betriebssystem aus unseren Dateimanager https://www.onworks.net/myfiles.php?username=XXXXX mit dem gewünschten Benutzernamen auf.
- 6. Laden Sie die Anwendung herunter, installieren Sie sie und führen Sie sie aus.
SCREENSHOTS:
Modellsuche
BESCHREIBUNG:
Model Search is an AutoML research system for discovering neural network architectures with minimal human intervention. Instead of hand-crafting models, you define a search space and objectives, then the system explores candidate architectures using controllers and population-based strategies. It supports multiple tasks (such as vision or text) by letting you express reusable building blocks—layers, cells, and topologies—that the search can recombine. Training, evaluation, and promotion of candidates are orchestrated automatically, with strong emphasis on reproducibility and fair comparisons. The framework logs trials, metrics, and artifacts so you can analyze what the search learned and why certain designs dominate. It’s intended as a platform for method development as much as for model discovery.
Eigenschaften
- Declarative search spaces for layers, cells, and connections
- Population-based training with promotion, mutation, and exploitation
- Multi-task support with pluggable input pipelines and heads
- Automated trial management, checkpointing, and early stopping
- Rich metric logging and artifact tracking for post-hoc analysis
- Extensible controllers to test new NAS algorithms
Programmiersprache
Python
Kategorien
This is an application that can also be fetched from https://sourceforge.net/projects/model-search.mirror/. It has been hosted in OnWorks in order to be run online in an easiest way from one of our free Operative Systems.