nni
nni copied to clipboard
microsoft
An open source AutoML toolkit for automate machine learning lifecycle, including feature engineering, neural architecture search, model compression and hyper-parameter tuning.
Release manager: @Bonytu Release start date: 5.15 Feature freeze date (at most 30 days): 6.12 Code freeze date & first package: Tutorial freeze: Release date (3 weeks since feature freeze):
Hello everyone, I would like to test ProxylessNAS on a smaller dataset like CIFAR using NNI version 3.0. I used the code that inherits from the DARTS example. However, it...
**Describe the issue**: running the following example experiment: `nnictl create --config .\nni\examples\trials\mnist-pytorch\config_windows.yml` getting this error: [2023-08-18 10:22:56] Creating experiment, Experiment ID: nu5kfm0o [2023-08-18 10:22:56] Starting web server... [2023-08-18 10:22:59] WARNING:...
**Describe the issue**: When I use "model_space = ProxylessNAS()" as the model_space of RetiariiExperiment, there is an error: TypeError: Repeat is not supported  **Environment**: - NNI version:2.10.1 - Training...
**Describe the issue**: Hello, I applied the L1NormPruner as in the tutorial (to all Conv2d layers in my model) but the accuracy drops from ~50 (mIOU) to 0 and it...
**Describe the issue**: In version 3.0rc1 TPE does not seem to be compatible with the InputChoice primitive. In addition, I found TPE tuner is default to minimize (I'm assuming because...
I am able to run FBNet (https://nni.readthedocs.io/en/v2.5/NAS/FBNet.html) when I used NNI version smaller than 2.9. However, for the master branch, there has been a refactoring, which has led to the...
**Describe the issue**: filename : examples/tutorials/hello_nas.py code : from nni.nas.nn.pytorch import LayerChoice, ModelSpace, MutableDropout, MutableLinear error : can't find MutableDropout, MutableLinear **Environment**: - NNI version:2.10.1 - Training service (local|remote|pai|aml|etc):local -...
I get a problem,I try to run my program with NNI Command "nnictl create --config config.yml --port 60001 --timestamp 88",and the trail-command in config.yml is "python framework.py --timestrap %timestrap%".
lib/python3.8/site-packages/nni/compression/pytorch/speedup/jit_translate.py", line 226, in __call__ assert len(args) >= len(self.undetermined) AssertionError
