🐛 Bug

I stumbled upon excessive CPU usage for my training code running on GPU. After some investigations I found the culprit.
It basically was

x = torch.eye(256).to('cuda') 

To Reproduce

This is quick and loads single CPU core.

%%timeit
    torch.eye(181)
6.43 µs ± 218 ns per loop (mean ± std. dev. of 7 runs, 100000 loops each)

This is 3 times slowe

At this moment relu_layer op doesn't allow threshold configuration, and legacy RELU op allows that.
We should add configuration option to relu_layer.

Problem: the approximate method can still be slow for many trees
catboost version: master
Operating System: ubuntu 18.04
CPU: i9
GPU: RTX2080

Would be good to be able to specify how many trees to use for shapley. The model.predict and prediction_type versions allow this. lgbm/xgb allow this.

As seen in openwall/john#4530 (comment):

Benchmarking: sspr-opencl, NetIQ SSPR / Adobe AEM [MD5/SHA1/SHA2 OpenCL]... Warning: binary() returned misaligned pointer
DONE

This is because opencl_sspr_fmt_plug.c wrongly has:

#define BINARY_ALIGN            MEM_ALIGN_WORD

whereas the code only guarantees alignment appropriate for

Hi ,

I have tried out both loss.backward() and model_engine.backward(loss) for my code. There are several subtle differences that I have observed , for one retain_graph = True does not work for model_engine.backward(loss) . This is creating a problem since buffers are not being retained every time I run the code for some reason.

Please look into this if you could.

Is your feature request related to a problem? Please describe.
It might be useful to have a singular clean and performant way to check if all the columns of a dataframe are of the same dtype, such as a DataFrame property _is_homogeneous. This comes up in a lot of places, such as where we might want to dispatch to a cupy matrix implementation (Transpose, some row wise reductions I believe

Current implementation of join can be improved by performing the operation in a single call to the backend kernel instead of multiple calls.

This is a fairly easy kernel and may be a good issue for someone getting to know CUDA/ArrayFire internals. Ping me if you want additional info.

We would like to forward a particular 'key' column which is part of the features to appear alongside the predictions - this is to be able to identify to which set of features a particular prediction belongs to. Here is an example of predictions output using the tensorflow.contrib.estimator.multi_class_head:

{"classes": ["0", "1", "2", "3", "4", "5", "6", "7", "8", "9"],
 "scores": [0.068196

Names map and input are exchanged mistakenly. By sense of Preconditions paragraph they have to be exchanged I suppose, because there is no problem when map and result coincide (in current context).