Avyav Kumar Singh

For branch v0.1, I want to get the complete pre-trained model without task-specific layer from the checkpoint file `mt_dnn_base_uncased.pt`, however, I am not able to convert it to `MTDNNModel`. Is...

Clarification of behaviour for recursive line generation

1

Hi, While generating lines for 768 dimensional data with a large number of classes (1609), the following lines were gotten using `find_lines_R_multiD` using k=~800 ``` [array([2, 2]) array([581, 780, 275])...

Errors while computing lines using recursive regression

11

Hi, While running the algorithm for generating lines using recursive regression, I noticed the following exceptions - ``` R[write to console]: Error in xtx_in %*% t(v) : non-conformable arguments Traceback...

Confirmation of expected behaviour of mathutils.geometry.intersect_point_line for higher dimensions

1

From the documentation of [mathutils.geometry.intersect_point_line](https://docs.blender.org/api/blender_python_api_2_78_release/mathutils.geometry.html#mathutils.geometry.intersect_point_line) the point closest to the given point on a line should be returned. it looks like the method does not work for higher dimensions -...

Unable to find lines using recursive_regression for higher dimensions

2

I was experimenting with the experiments, specifically trying to generate lines for data points with high dimensions. I was able to reproduce the error on [this notebook](https://github.com/ilia10000/LO-Shot/blob/master/Paper3/Soft%20Label%20Optimization.ipynb). The snippet can...

Hi, From https://github.com/iesl/leopard, it looks like data for K=32 is missing. Is there a workaround for this issue? When I try to fine-tune the models, I get this result -...