QuanluZhang

the following updates to the source code can fix this bug. in the file https://github.com/microsoft/nni/blob/master/nni/compression/pytorch/quantization_speedup/frontend_to_onnx.py#L127 change the line 127, 128 from ```python index2name[i] = name name2index[name] = i ``` to...

@Raychen0617 , thanks for reporting this issue, which tutorial did you run?

A related issue for your reference #4869 , in where the provided solution also does not support reporting intermediate result. The reason has been explained by @ultmaster. Currently, one hacky...

@juliuskittler thanks for your good suggestion. NNI allows user to do model selection by using hyper-parameter tuning. [Here](https://github.com/microsoft/nni/blob/master/examples/trials/sklearn/regression/search_space.json) is a very simple example, and users can further leverage [nested search...

@xiaoerqi thanks for raising this issue. Directly calling a train() function in script is indeed convenient in some scenarios, will support this feature probably in v3.1 release.

The next release is v2.9. We have planned items for v2.9 and v3.0. Therefore, this feature is planned in v3.1

@sunpeil thanks for raising this issue. The current implementation of DoReFaQuantizer only supports quantization of weights. The original paper also supports quantization of activation tensors. The support of activation quantization...

- [ ] double check stdout stderr (log) of trial in remote mode and all other training services - [ ] whether all trials' log are written in one log...

@impulsecorp thanks for raising this issue. The summary of the problem is very clear. We will support view and resume of Retiarii experiment in v2.9. For the upcoming v2.8, you...

First, for resume/view, you can use either python API (e.g., RetiariiExperiment.resume(...), RetiariiExperiment.view(...)) or nnictl, which will be supported in v2.9. Second, after the best model is exported (it is a...