Uri Cohen

Tensor dimensions reversed?

5

In matlab, I write a single 1925x4 variable into a mat file: >> size(params) 1925 4 >> save('file.mat', 'params') Loading it into torch result with the dimensions reversed: th> require...

Handling the case of changed default tensor type (e.g. overfeat torch do the following: torch.setdefaulttensortype('torch.FloatTensor') which break the code as torch.Tensor():resizeAs() would expect a FloatTensor.

The [documentation](https://allensdk.readthedocs.io/en/latest/visual_coding_neuropixels.html) states that "Every NWB file includes a table of quality metrics" and that "By default, we won’t show you units below a pre-determined quality threshold; we hide any...

ecos crash on a very sparse problem

1

The code below reproduces a non-deterministic bug which causes `ecos` to crash with the following message, on both Windows (python 3.10) and Gentoo Linux (python 3.8). Despite using the same...

Updates to ModelDB.xml are not deterministic

2

If ModelDB.xml is updated by different people (using bin/neuronvisio-modeldb-updater) the result may differ. This is due to non-deterministic of the ordering of Python's dictionary.keys() used to insert new items, as...

Unuable to remove the last layers of a model

2

I am using the caffe reference network as a feature-extractor and would like to remove the last softmax layers (and linear layer `fc8`). It seems that editing the `deploy.prototxt` file...

Problems using converted model

2

Hello, I tried to use a model I trained on pytorch from matconvnet using your wonderful project, but something is not right with the results I get from the converted...

python2 support

1

Hello, Can the python3 dependency avoided? I use pytorch on a cluster where I have it installed with python2.7, and I could not install it with python3 (due to pytorch...