docs: moving three tutorials from Pruna Pro to Pruna

docs/tutorials/computer_vision.ipynb

@@ -0,0 +1,177 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Blazingly fast Computer Vision Models"
+   ]
+  },
+  {
+   "cell_type": "raw",
+   "metadata": {
+    "vscode": {
+     "languageId": "raw"
+    }
+   },
+   "source": [
+    "<a target=\"_blank\" href=\"https://colab.research.google.com/drive/1GkzxTQW-2yCKXc8omE6Sa4SxiETMi8yC?usp=sharing\\\">\n",
+    "    <img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/>\n",
+    "</a>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "This tutorial demonstrates how to use the `pruna` package to optimize any custom computer vision model. We will use the `vit_b_16` model as an example. Any execution times given below are measured on a T4 GPU."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### 1. Loading the CV Model\n",
+    "\n",
+    "First, load your ViT model.\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import torchvision\n",
+    "\n",
+    "model = torchvision.models.vit_b_16(weights=\"ViT_B_16_Weights.DEFAULT\").cuda()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### 2. Initializing the Smash Config\n",
+    "\n",
+    "Next, initialize the smash_config."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from pruna import SmashConfig\n",
+    "\n",
+    "# Initialize the SmashConfig\n",
+    "smash_config = SmashConfig([\"x_fast\"])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### 3. Smashing the Model\n",
+    "\n",
+    "Now, you can smash the model, which will take around 5 seconds."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from pruna import smash\n",
+    "\n",
+    "# Smash the model\n",
+    "smashed_model = smash(\n",
+    "    model=model,\n",
+    "    smash_config=smash_config,\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### 4. Preparing the Input"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "from torchvision import transforms\n",
+    "\n",
+    "# Generating a random image\n",
+    "image = np.random.randint(0, 256, size=(224, 224, 3), dtype=np.uint8)\n",
+    "input_tensor = transforms.ToTensor()(image).unsqueeze(0).cuda()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### 5. Running the Model\n",
+    "\n",
+    "After the model has been compiled, we run inference for a few iterations as warm-up. This will take around 8 seconds."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# run some warm-up iterations\n",
+    "for _ in range(5):\n",
+    "  smashed_model(input_tensor)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Finally, run the model to transcribe the audio file with accelerated inference."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Display the result\n",
+    "smashed_model(input_tensor)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Wrap Up\n",
+    "\n",
+    "Congratulations! You have successfully smashed a CV model. You can now use the `pruna` package to optimize any custom CV model. The only parts that you should modify are step 1, 4 and 5 to fit your use case"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "pruna",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "name": "python",
+   "version": "3.10.15"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}

docs/tutorials/index.rst

@@ -87,10 +87,44 @@ These tutorials will guide you through the process of using |pruna| to optimize
 
       Learn how to use the ``target_modules`` parameter to target specific modules in your model.
 
+   .. grid-item-card:: Blazingly Fast Computer Vision
+      :text-align: center
+      :link: ./computer_vision.ipynb
+
+      Optimize any ``computer vision`` model with ``x_fast`` ``compilation``.
+
+   .. grid-item-card:: Recover Quality after Quantization
+      :text-align: center
+      :link: ./recovery.ipynb
+
+      Recover quality using ``text_to_image_perp`` after ``diffusers_int8`` ``quantization``.
+
+   .. grid-item-card:: Distribute across GPUs with Ring Attention
+      :text-align: center
+      :link: ./ring_attn.ipynb
+
+      Distribute your ``Flux`` model across multiple GPUs with ``ring_attn`` and ``torch_compile``.
+
 .. toctree::
    :hidden:
    :maxdepth: 1
    :caption: Pruna
-   :glob:
-
-   ./*
+
+   image_generation
+   video_generation
+   llms
+   reasoning_llm
+   asr_tutorial
+   cv_cpu
+   diffusion_quantization_acceleration
+   evaluation_agent_cmmd
+   sana_diffusers_int8
+   flux2klein4b_tutorial
+   sd_deepcache
+   deploying_sana_tutorial
+   target_modules_quanto
+   portable_compilation
+   llm_quantization_compilation_acceleration
+   computer_vision
+   recovery
+   ring_attn