Nick Rossenbach
Nick Rossenbach
When running the `compile_native_op.sh` you will always get the BLAS lib from numpy linked into the binary. - When not adding any flag, it will load the BLAS lib from...
Removing the `-std=...` flag as default in #896 caused the breaking of many setups, as the i6 environment is not properly reflected by the automatic test. Before we continue altering...
I would like to refactor the datasets code in a way, that the package/module structure is actually more reasonable. This is especially about cleaning up `generating.py`, as more than 50%...
I know that we do not want to do it right away, but I still wanted to make an issue about this, as I would be in favor of doing...
Discussion related to #508 using `num_inputs` and `num_outputs` should throw an error message as new behaviour. Only `extern_data` should be used.
#508 This is old Theano behavior.
Discussion related to #508 Many parameters that are part of the dataset can be set globally: ```python3 set_or_remove("window", config.int('window', 0) or None) set_or_remove("context_window", config.typed_value("context_window")) set_or_remove("chunking", config.opt_typed_value("chunking", None)) set_or_remove("seq_ordering", config.value("batching", None))...
Discussion related to #508 Right now there are both a `learning_rate` and a `learning_rates` parameter, which often causes confusion to the users. Here I am not sure what would be...
It seems to me that the `HDFDataset` does only support data streams with a single time-axis, and not data with no or multiple time axes. I know that I can...
When `dropout_noise_shape` is `None`, it will automatically broadcast the dropout values across the batch dimension and all spatial axes to save computing time. Currently, there is no way to set...