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.
**Hello guys, i am using PolicyBasedRL strategy to do multi trial NAS experiments, when I set concurrency=2, the tuner crased, and concurrency=1 works just fine. Here is the error message**:...
Error message is {"error":"Setup training service failed."} ERROR: Failed! Error is: None   i meet bug like this when i try the trial in examples/nas/classic_nas but i have sucessfully...
**Describe the issue**:When experient is over, and I didnt open it in bowser at fisrt, so how to check the experiments in browser ?
**Describe the issue**: when i try to speed up my model, i meet such a problem: /home/user/.conda/envs/yang_pytorch/lib/python3.7/site-packages/nni/compression/pytorch/utils/mask_conflict.py:124: UserWarning: This overload of nonzero is deprecated: nonzero() Consider using one of the...
 Link here should be `https://github.com/Microsoft/nni/tree/v2.1` instead of `https://github.com/Microsoft/nni/tree/2.1.0`
**Describe the issue**: **Cannot speed up depthwise separable convolutions** [2022-05-16 11:22:40] INFO (nni.compression.pytorch.speedup.compressor/MainThread) replace module (name: _backbone._block2.1._dw_conv.0, op_type: Conv2d) Traceback (most recent call last): File "D:/nni-master/examples/model_compress/pruning/FD_Prune/FD_simulated_prune.py", line 278, in pruner.compress()...
**Environment**: - NNI version: 2.1 - Python version: 3.6.9 (default version in the image) - PyTorch/TensorFlow version: Tensorflow 2.3.1 (default version in the image) - Is conda/virtualenv/venv used?: no -...
Currently, it's hard to know status of tuners, assessors, and advisors. Log can help it, but the log of them are mixed within nnimanager, and hard to have a overall...
**Describe the issue**: Hi there, just starting to use `nni` and was interested in trying some features in the master branch, but couldn't get the install to work. The error...
