Kevin Mader

Results 97 issues of Kevin Mader

adding an example to get the path (https://github.com/peteut/ipython-file-upload/issues/18) to make it easier to open an uploaded file with tools that don't support buffered IO directly.

add more detail to weight loading errors to help debugging loading [issues](https://github.com/transcranial/keras-js/issues/60) Updated error message is more useful for fixing the model ``` Error: [Model] error loading weights prebn_elbow/gamma,prebn_elbow/beta,prebn_elbow/moving_mean,prebn_elbow/moving_variancefor BatchNormalization...

I have tried to use a nested model and get a JavaScript exception when loading the model ``` Error: [Model] Layer Sequential specified in model configuration is not implemented! ```...

I have setup a test environment with libeigen3-dev installed (/usr/include/eigen3) The webassembly block https://github.com/mil-tokyo/webdnn/blob/afe16593463b3ee9519d285baf693547fd5f25f9/src/graph_transpiler/webdnn/backend/webassembly/generator.py#L78-L99 does not find eigen and reports ``` BackendName: webassembly Executing em++ command failed. Make sure emscripten...

adding link to eigen3 path to hopefully address this [issue](https://github.com/mil-tokyo/webdnn/issues/777)

adding h5py requirement since that is not included in base keras

Performance on fairly simple 1D model with stacked branches / bottlenecks / inception blocks was surprisingly poor (>300x slower). ## Keras on CPU (tensorflow lite for the same model is...

help wanted

When comparing to random sampling it does not seem to give significantly different results. I would have expected the curve to be much higher for active learning. Potentially the defaults...

I started to implement some basic testing for the notebooks in response to the comment here: https://github.com/tensorflow/lucid/pull/97#issuecomment-411925091

awaiting reply

Perhaps this is a rare use-case, but I often have folders full of .ply files to load into one cloud. I am currently manually reading them in (just points, probably...

Feature Request
Debate