Anwaar Khalid

Handling <unsupported/supported> <dypes/devices> for mixed functions

14

- [Trello: Fix `recurse` option in `ivy.function_supported_devices_and_dtypes` and `ivy.function_unsupported_devices_and_dtypes`](https://trello.com/c/cjfEWe4b/1668-fix-recurse-option-in-ivyfunctionsupporteddevicesanddtypes-and-ivyfunctionunsupporteddevicesanddtypes) - [Trello: Investigate why `_get_supported_devices_dtypes` function is not returning the correct dtype set for mixed functions](https://trello.com/c/8IQK5r9V/1649-investigate-why-getsupporteddevicesdtypes-function-is-not-returning-the-correct-dtype-set-for-mixed-functions)

Hey @debidatta and RepNet team, hope you are all doing well. First of all, thanks for this super cool model and the detailed [collab](https://github.com/google-research/google-research/blob/master/repnet/repnet_colab.ipynb) notebook :fire: .I've been exploring the...

Randomised_CP function throws a Singular Matrix error

4

#### Steps or Code to Reproduce ```python import tensorly as tl import ivy tl.set_backend("numpy") a = tl.ones((2,3,4)) z = tl.decomposition.randomised_parafac(a, 2, 1, n_iter_max=1, init="svd", max_stagnation=1, tol=0.03125, random_state=1) print(z) ``` ####...

`tltorch.FactorizedConv.from_conv` tries to allocate ~83GB memory for an input of shape (1024, 512, 3, 3, 3)

2

Minimal Code to reproduce the error: ```python import torch import tltorch test_conv3d = torch.nn.Conv3d(1024, 512, (3,3,3), padding=(3,1,1)) print(tltorch.FactorizedConv.from_conv(test_conv3d, rank='same', factorization='cp')) ``` Error: ``` RuntimeError: [enforce fail at alloc_cpu.cpp:83] err ==...

`torch.jit.script` does not work with Tensorized Models

1

Minimal Code: ```python import torch from torch.nn import Module from tltorch import FactorizedConv class Test(Module): def __init__(self): super(Test, self).__init__() self.layer = FactorizedConv(3, 4, 3, factorization='tucker', order=3) def main(): # Instantiate...

https://trello.com/c/qVgTiCNH/1673-yolo-demo-and-implementation Creating this PR so that the work done on this model so far doesn't get lost. Since this model is not required for any demo, I've stopped working on...

I'm working with a text to speech model so I cannot predict the size of the output tensors beforehand. According to the [documentation](https://www.tensorflow.org/lite/guide/inference#outputs), "some models have dynamic outputs, where the...

I was comparing the rotary embedding implementation in this repository with the implementations in the official Llama and Deepseek repositories using this Jupyter notebook: [link](https://colab.research.google.com/drive/1I9aBN55UUgmUwSNTmELC1u7DWuEk1dU2?usp=sharing). In Llama and Deepseek repositories,...

Add HQQ to weight compression algorithms for LLMs

1

### 🚀 Feature request [HQQ](https://github.com/mobiusml/hqq) is a popular data-free weight quantization algorithm for LLMs. It would be super cool to add it NNCF's weight compression algorithms. I would like to...