sparseml/integrations update #1408 broke links in docs

[evankiely]

What is the URL, file, or UI containing proposed doc change
I recognize that this section is requesting some set of changes I've attempted to enact for your reference, but I don't actually have a comprehensive set of changes at this point (though see section "What is the proposed change" for my attempt). Instead, please reference this page, which contains the now broken links. There are other pages like this with similar issues (see below for one example), but I've not attempted to identify them all. I should also note that I am presuming it was update #1408 which caused this issue, as I haven't explored the docs much prior to this update, so I cannot confirm at what point the links broke. However, it does seem like this is the update which moved things around (for example).

Here is another example of a page which features some of the same broken links.

What is the current content or situation in question
All of the following return 404 error (this is not intended to be a comprehensive accounting of all links which do):

Tutorials
Computer Vision
Sparsifying PyTorch Models Using Recipes
Sparsifying YOLOv3 Using Recipes
Sparsifying YOLOv5 Using Recipes
Sparsifying YOLACT Using Recipes
Sparse Transfer Learning for Image Classification
Sparse Transfer Learning With YOLOv3
Sparse Transfer Learning With YOLOv5

Errors on 404 page (screenshots provided as well):

• Sparsifying PyTorch Models Using Recipes
  • 
  • "The 'neuralmagic/sparseml' repository doesn't contain the 'integrations/pytorch/tutorials/sparsifying_pytorch_models_using_recipes.md' path in 'main'."
• Sparsifying YOLOv3 Using Recipes
  • 
  • "The 'neuralmagic/sparseml' repository doesn't contain the 'integrations/ultralytics-yolov3/tutorials/sparsifying_yolov3_using_recipes.md' path in 'main'."
• Sparsifying YOLOv5 Using Recipes
  • 
  • The 'neuralmagic/sparseml' repository doesn't contain the 'integrations/ultralytics-yolov5/tutorials/sparsifying_yolov5_using_recipes.md' path in 'main'.
• Sparsifying YOLACT Using Recipes
  • 
  • The 'neuralmagic/sparseml' repository doesn't contain the 'integrations/dbolya-yolact/tutorials/sparsifying_yolact_using_recipes.md' path in 'main'.
• Sparse Transfer Learning for Image Classification
  • 
  • The 'neuralmagic/sparseml' repository doesn't contain the 'integrations/pytorch/tutorials/classification_sparse_transfer_learning_tutorial.md' path in 'main'.
• Sparse Transfer Learning With YOLOv3
  • 
  • The 'neuralmagic/sparseml' repository doesn't contain the 'integrations/pytorch/tutorials/classification_sparse_transfer_learning_tutorial.md' path in 'main'.
• Sparse Transfer Learning With YOLOv5
  • 
  • The 'neuralmagic/sparseml' repository doesn't contain the 'integrations/ultralytics-yolov3/tutorials/yolov3_sparse_transfer_learning.md' path in 'main'.

What is the proposed change
While it is tempting to simply point these links at the docs they once referenced, but which have been moved elsewhere due to being old/out of date, it's probably best to, instead, remap these to the proper links to updated versions -- though this is easier said than done, from what I gathered during my admittedly brief attempt. Alternatively, if there are new versions of the same docs, this may resolve itself once those are uploaded..?

I have attempted to find the mappings for the above and provided them below. These aren't necessarily valid mappings, but I've included an excerpt from each to demonstrate why I felt it might be an appropriate mapping, for those cases where I found a seemingly appropriate alternative at all.

• Sparsifying PyTorch Models Using Recipes
  • Couldn't identify a good alternative
• Sparsifying YOLOv3 Using Recipes
  • "This tutorial shows how Neural Magic recipes simplify the sparsification process by encoding the hyperparameters and instructions needed to create highly accurate pruned and pruned-quantized YOLOv3 models."
  • NOTE: This is in the "Old Examples" directory, so may just want to omit.
• Sparsifying YOLOv5 Using Recipes
  • "In this tutorial, we will demonstrate how to use recipes to create sparse versions of YOLOv5."
  • Also linked to from the main readme
• Sparsifying YOLACT Using Recipes
  • Couldn't identify a good alternative
• Sparse Transfer Learning for Image Classification
  • Titled "Sparsifying Image Classification Models with the CLI"
  • First sub-title "Sparse Transfer Learning Overview"
  • Also linked to from the main readme
• Sparse Transfer Learning With YOLOv3
  • Titled: "Sparse Transfer Learning With YOLOv3"
  • NOTE: This is in the "Old Examples" directory, so may just want to omit.
• Sparse Transfer Learning With YOLOv5
  • "However, with Sparse Transfer Learning, we simply start the fine-tuning process from a pre-sparsified YOLOv5 and maintain sparsity while the training process occurs."
  • Also linked to from the main readme

Additional context
N/A but willing to help however I can.
P.s. Thanks for your patience, I'm still new to submitting bug reports, etc., and not even sure this is the place to log an issue such as this one.

[mgoin]

@evankiely Thank you for the thorough report on these URL issues! We are going through some large restructuring this release cycle with our documentation on SparseML and DeepSparse. We will work quickly to fix these bad links

[evankiely]

@mgoin But of course, and thank you all for such an interesting and useful suite of tools! Please let me know if I can be of any further help 🙂

[jeanniefinks]

Per the main README announcement, SparseML is being deprecated by early June 2, 2025. Closing the issue and thank you for the inputs and support! We appreciate it!