diffusion/diffusers/generating_by_fp16.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Generating an image by fp16 "
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This process took ~30s on a T4 GPU(from official document), 43.2s from M1 GPU. By default, the DiffusionPipeline runs inference with full float32 precision for 50 inference steps. We can speed up by switching to a lower precision like float16 or running fewer inference steps. But it cannot support on CPU."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preparing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "!pip install diffusers[\"torch\",'flax'] transformers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os, platform\n",
    "\n",
    "torch_device = 'cpu'\n",
    "\n",
    "if 'kaggle' in os.environ.get('KAGGLE_URL_BASE','localhost'):\n",
    "    torch_device = 'cuda'\n",
    "else:\n",
    "    torch_device = 'mps' if platform.system() == 'Darwin' else 'cpu'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'mps'"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "torch_device"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "os.environ['PYTORCH_ENABLE_MPS_FALLBACK'] = '1'"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loading models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "from diffusers import DiffusionPipeline\n",
    "\n",
    "\n",
    "model_id=\"runwayml/stable-diffusion-v1-5\"\n",
    "pipeline = DiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.float16)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "prompt = \"portrait photo of master sword in Zelda: Breath of the Wild\""
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Generating"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pipeline = pipeline.to(torch_device)\n",
    "generator = torch.Generator(device=torch_device).manual_seed(0)\n",
    "image = pipeline(prompt, generator=generator).images[0]\n",
    "image"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "diffusers",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.3"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}