Results.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Results\n",
    "Here, we will inspect the already trained model and its capabilities. However, we will not be evaluating performance here.\n",
    "\n",
    "### List of Contents\n",
    "- [Set Global Params](#set-global-params)\n",
    "- [Inspect Model](#inspect-model)\n",
    "    - Model Summary\n",
    "    - Test input/output\n",
    "    - Sample Images\n",
    "    - Denoise Images\n",
    "    - Inspect Filters\n",
    "- [Generate Rotational Images](#generate-rotational-images)\n",
    "- [Misc. functionalities](#misc-functionalities)\n",
    "    - Save MNIST dataset from CSV to PNG images\n",
    "    - Make Collage of Training Images\n",
    "\n",
    "\n",
    "First, let's import the libraries."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "from modules.utils import *\n",
    "from modules.filtrs import *\n",
    "from modules.utils import *\n",
    "from modules.ddpm_utils import *\n",
    "from modules.ddpm_models import *\n",
    "from modules.ddpm_tasks import rotation_results,make_video\n",
    "import pickle"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Set Global Params"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# global \n",
    "random_seed=42\n",
    "args = argument()\n",
    "args.image_size = 32\n",
    "args.image_channels=1#1\n",
    "args.device = \"cuda\"\n",
    "args.noise_steps=1000"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Inspect Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Variant 3 Modified UNet: filters around gelu + filters in up or downsampling\n",
      "Unconditional UNet\n"
     ]
    }
   ],
   "source": [
    "args=args\n",
    "unet_variant=3\n",
    "f_settings={}\n",
    "f_settings['kernel_size']=3\n",
    "f_settings['kaiser_beta']=2\n",
    "f_settings['omega_c_down'] =np.pi/2\n",
    "f_settings['omega_c_up'] = np.pi/2\n",
    "\n",
    "modelpath=\"E:\\Research\\Diffusion-Models-pytorch\\models\\DDPM_Uncondtional_MNIST_3\\ckpt_MNIST_3.pt\"\n",
    "\n",
    "set_seed(random_seed)\n",
    "model = UNet(c_in=args.image_channels, c_out=args.image_channels,\n",
    "            image_size=args.image_size,f_settings=f_settings,device=args.device,variant=unet_variant).to(args.device)\n",
    "ckpt = torch.load(modelpath)\n",
    "model.load_state_dict(ckpt)\n",
    "diffusion = Diffusion(noise_steps=args.noise_steps,img_size=args.image_size, device=args.device)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model summary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5896513\n",
      "UNet(\n",
      "  (inc): DoubleConv_F(\n",
      "    (conv1): Conv2d(1, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "    (norm1): GroupNorm(1, 32, eps=1e-05, affine=True)\n",
      "    (gelu): GELU(approximate='none')\n",
      "    (conv2): Conv2d(32, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "    (norm2): GroupNorm(1, 32, eps=1e-05, affine=True)\n",
      "  )\n",
      "  (down1): Down_FFF(\n",
      "    (conv): Sequential(\n",
      "      (0): DoubleConv_F(\n",
      "        (conv1): Conv2d(32, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (norm1): GroupNorm(1, 32, eps=1e-05, affine=True)\n",
      "        (gelu): GELU(approximate='none')\n",
      "        (conv2): Conv2d(32, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (norm2): GroupNorm(1, 32, eps=1e-05, affine=True)\n",
      "      )\n",
      "      (1): DoubleConv_F(\n",
      "        (conv1): Conv2d(32, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (norm1): GroupNorm(1, 64, eps=1e-05, affine=True)\n",
      "        (gelu): GELU(approximate='none')\n",
      "        (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (norm2): GroupNorm(1, 64, eps=1e-05, affine=True)\n",
      "      )\n",
      "    )\n",
      "    (emb_layer): Sequential(\n",
      "      (0): SiLU()\n",
      "      (1): Linear(in_features=256, out_features=64, bias=True)\n",
      "    )\n",
      "  )\n",
      "  (sa1): SelfAttention(\n",
      "    (mha): MultiheadAttention(\n",
      "      (out_proj): NonDynamicallyQuantizableLinear(in_features=64, out_features=64, bias=True)\n",
      "    )\n",
      "    (ln): LayerNorm((64,), eps=1e-05, elementwise_affine=True)\n",
      "    (ff_self): Sequential(\n",
      "      (0): LayerNorm((64,), eps=1e-05, elementwise_affine=True)\n",
      "      (1): Linear(in_features=64, out_features=64, bias=True)\n",
      "      (2): GELU(approximate='none')\n",
      "      (3): Linear(in_features=64, out_features=64, bias=True)\n",
      "    )\n",
      "  )\n",
      "  (down2): Down_FFF(\n",
      "    (conv): Sequential(\n",
      "      (0): DoubleConv_F(\n",
      "        (conv1): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (norm1): GroupNorm(1, 64, eps=1e-05, affine=True)\n",
      "        (gelu): GELU(approximate='none')\n",
      "        (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (norm2): GroupNorm(1, 64, eps=1e-05, affine=True)\n",
      "      )\n",
      "      (1): DoubleConv_F(\n",
      "        (conv1): Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (norm1): GroupNorm(1, 128, eps=1e-05, affine=True)\n",
      "        (gelu): GELU(approximate='none')\n",
      "        (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (norm2): GroupNorm(1, 128, eps=1e-05, affine=True)\n",
      "      )\n",
      "    )\n",
      "    (emb_layer): Sequential(\n",
      "      (0): SiLU()\n",
      "      (1): Linear(in_features=256, out_features=128, bias=True)\n",
      "    )\n",
      "  )\n",
      "  (sa2): SelfAttention(\n",
      "    (mha): MultiheadAttention(\n",
      "      (out_proj): NonDynamicallyQuantizableLinear(in_features=128, out_features=128, bias=True)\n",
      "    )\n",
      "    (ln): LayerNorm((128,), eps=1e-05, elementwise_affine=True)\n",
      "    (ff_self): Sequential(\n",
      "      (0): LayerNorm((128,), eps=1e-05, elementwise_affine=True)\n",
      "      (1): Linear(in_features=128, out_features=128, bias=True)\n",
      "      (2): GELU(approximate='none')\n",
      "      (3): Linear(in_features=128, out_features=128, bias=True)\n",
      "    )\n",
      "  )\n",
      "  (down3): Down_FFF(\n",
      "    (conv): Sequential(\n",
      "      (0): DoubleConv_F(\n",
      "        (conv1): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (norm1): GroupNorm(1, 128, eps=1e-05, affine=True)\n",
      "        (gelu): GELU(approximate='none')\n",
      "        (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (norm2): GroupNorm(1, 128, eps=1e-05, affine=True)\n",
      "      )\n",
      "      (1): DoubleConv_F(\n",
      "        (conv1): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (norm1): GroupNorm(1, 128, eps=1e-05, affine=True)\n",
      "        (gelu): GELU(approximate='none')\n",
      "        (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (norm2): GroupNorm(1, 128, eps=1e-05, affine=True)\n",
      "      )\n",
      "    )\n",
      "    (emb_layer): Sequential(\n",
      "      (0): SiLU()\n",
      "      (1): Linear(in_features=256, out_features=128, bias=True)\n",
      "    )\n",
      "  )\n",
      "  (sa3): SelfAttention(\n",
      "    (mha): MultiheadAttention(\n",
      "      (out_proj): NonDynamicallyQuantizableLinear(in_features=128, out_features=128, bias=True)\n",
      "    )\n",
      "    (ln): LayerNorm((128,), eps=1e-05, elementwise_affine=True)\n",
      "    (ff_self): Sequential(\n",
      "      (0): LayerNorm((128,), eps=1e-05, elementwise_affine=True)\n",
      "      (1): Linear(in_features=128, out_features=128, bias=True)\n",
      "      (2): GELU(approximate='none')\n",
      "      (3): Linear(in_features=128, out_features=128, bias=True)\n",
      "    )\n",
      "  )\n",
      "  (bot1): DoubleConv_F(\n",
      "    (conv1): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "    (norm1): GroupNorm(1, 256, eps=1e-05, affine=True)\n",
      "    (gelu): GELU(approximate='none')\n",
      "    (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "    (norm2): GroupNorm(1, 256, eps=1e-05, affine=True)\n",
      "  )\n",
      "  (bot2): DoubleConv_F(\n",
      "    (conv1): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "    (norm1): GroupNorm(1, 256, eps=1e-05, affine=True)\n",
      "    (gelu): GELU(approximate='none')\n",
      "    (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "    (norm2): GroupNorm(1, 256, eps=1e-05, affine=True)\n",
      "  )\n",
      "  (bot3): DoubleConv_F(\n",
      "    (conv1): Conv2d(256, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "    (norm1): GroupNorm(1, 128, eps=1e-05, affine=True)\n",
      "    (gelu): GELU(approximate='none')\n",
      "    (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "    (norm2): GroupNorm(1, 128, eps=1e-05, affine=True)\n",
      "  )\n",
      "  (up1): Up_FFF(\n",
      "    (conv): Sequential(\n",
      "      (0): DoubleConv_F(\n",
      "        (conv1): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (norm1): GroupNorm(1, 256, eps=1e-05, affine=True)\n",
      "        (gelu): GELU(approximate='none')\n",
      "        (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (norm2): GroupNorm(1, 256, eps=1e-05, affine=True)\n",
      "      )\n",
      "      (1): DoubleConv_F(\n",
      "        (conv1): Conv2d(256, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (norm1): GroupNorm(1, 128, eps=1e-05, affine=True)\n",
      "        (gelu): GELU(approximate='none')\n",
      "        (conv2): Conv2d(128, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (norm2): GroupNorm(1, 64, eps=1e-05, affine=True)\n",
      "      )\n",
      "    )\n",
      "    (emb_layer): Sequential(\n",
      "      (0): SiLU()\n",
      "      (1): Linear(in_features=256, out_features=64, bias=True)\n",
      "    )\n",
      "  )\n",
      "  (sa4): SelfAttention(\n",
      "    (mha): MultiheadAttention(\n",
      "      (out_proj): NonDynamicallyQuantizableLinear(in_features=64, out_features=64, bias=True)\n",
      "    )\n",
      "    (ln): LayerNorm((64,), eps=1e-05, elementwise_affine=True)\n",
      "    (ff_self): Sequential(\n",
      "      (0): LayerNorm((64,), eps=1e-05, elementwise_affine=True)\n",
      "      (1): Linear(in_features=64, out_features=64, bias=True)\n",
      "      (2): GELU(approximate='none')\n",
      "      (3): Linear(in_features=64, out_features=64, bias=True)\n",
      "    )\n",
      "  )\n",
      "  (up2): Up_FFF(\n",
      "    (conv): Sequential(\n",
      "      (0): DoubleConv_F(\n",
      "        (conv1): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (norm1): GroupNorm(1, 128, eps=1e-05, affine=True)\n",
      "        (gelu): GELU(approximate='none')\n",
      "        (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (norm2): GroupNorm(1, 128, eps=1e-05, affine=True)\n",
      "      )\n",
      "      (1): DoubleConv_F(\n",
      "        (conv1): Conv2d(128, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (norm1): GroupNorm(1, 64, eps=1e-05, affine=True)\n",
      "        (gelu): GELU(approximate='none')\n",
      "        (conv2): Conv2d(64, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (norm2): GroupNorm(1, 32, eps=1e-05, affine=True)\n",
      "      )\n",
      "    )\n",
      "    (emb_layer): Sequential(\n",
      "      (0): SiLU()\n",
      "      (1): Linear(in_features=256, out_features=32, bias=True)\n",
      "    )\n",
      "  )\n",
      "  (sa5): SelfAttention(\n",
      "    (mha): MultiheadAttention(\n",
      "      (out_proj): NonDynamicallyQuantizableLinear(in_features=32, out_features=32, bias=True)\n",
      "    )\n",
      "    (ln): LayerNorm((32,), eps=1e-05, elementwise_affine=True)\n",
      "    (ff_self): Sequential(\n",
      "      (0): LayerNorm((32,), eps=1e-05, elementwise_affine=True)\n",
      "      (1): Linear(in_features=32, out_features=32, bias=True)\n",
      "      (2): GELU(approximate='none')\n",
      "      (3): Linear(in_features=32, out_features=32, bias=True)\n",
      "    )\n",
      "  )\n",
      "  (up3): Up_FFF(\n",
      "    (conv): Sequential(\n",
      "      (0): DoubleConv_F(\n",
      "        (conv1): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (norm1): GroupNorm(1, 64, eps=1e-05, affine=True)\n",
      "        (gelu): GELU(approximate='none')\n",
      "        (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (norm2): GroupNorm(1, 64, eps=1e-05, affine=True)\n",
      "      )\n",
      "      (1): DoubleConv_F(\n",
      "        (conv1): Conv2d(64, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (norm1): GroupNorm(1, 32, eps=1e-05, affine=True)\n",
      "        (gelu): GELU(approximate='none')\n",
      "        (conv2): Conv2d(32, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
      "        (norm2): GroupNorm(1, 32, eps=1e-05, affine=True)\n",
      "      )\n",
      "    )\n",
      "    (emb_layer): Sequential(\n",
      "      (0): SiLU()\n",
      "      (1): Linear(in_features=256, out_features=32, bias=True)\n",
      "    )\n",
      "  )\n",
      "  (sa6): SelfAttention(\n",
      "    (mha): MultiheadAttention(\n",
      "      (out_proj): NonDynamicallyQuantizableLinear(in_features=32, out_features=32, bias=True)\n",
      "    )\n",
      "    (ln): LayerNorm((32,), eps=1e-05, elementwise_affine=True)\n",
      "    (ff_self): Sequential(\n",
      "      (0): LayerNorm((32,), eps=1e-05, elementwise_affine=True)\n",
      "      (1): Linear(in_features=32, out_features=32, bias=True)\n",
      "      (2): GELU(approximate='none')\n",
      "      (3): Linear(in_features=32, out_features=32, bias=True)\n",
      "    )\n",
      "  )\n",
      "  (outc): Conv2d(32, 1, kernel_size=(1, 1), stride=(1, 1))\n",
      ")\n"
     ]
    }
   ],
   "source": [
    "print(sum([p.numel() for p in model.parameters()]))\n",
    "print(model)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Test input/output of the Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([2, 1, 32, 32])\n"
     ]
    }
   ],
   "source": [
    "x = torch.randn(2, args.image_channels, args.image_size, args.image_size).to(args.device)\n",
    "t = x.new_tensor([args.noise_steps] * x.shape[0]).long().to(args.device)\n",
    "y = x.new_tensor([1] * x.shape[0]).long().to(args.device)\n",
    "#print(net(x, t, y).shape) # conditional case\n",
    "print(model(x, t).shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Sample Images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "999it [00:31, 31.95it/s]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAACcQAAAGxCAYAAABGcycoAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/H5lhTAAAACXBIWXMAAA9hAAAPYQGoP6dpAAA9gklEQVR4nO3ce7BddXnw8bX3PrnfSQiXEC4m4R4lcjUSUEZBRSjoiKK2zFhtqY5jZ5zp2I51+kdbpjN2po4dsRfpTVOLl4BOhQxIuQiEKERSEspFCAkJCQm538/Ze79/vGM7rfPaPA+eH/v98fn8vb7nt/Y+a6+91sqT0+r3+/0GAAAAAAAAAAAA/j/Xfq13AAAAAAAAAAAAAH4VDMQBAAAAAAAAAABQBQNxAAAAAAAAAAAAVMFAHAAAAAAAAAAAAFUwEAcAAAAAAAAAAEAVDMQBAAAAAAAAAABQBQNxAAAAAAAAAAAAVMFAHAAAAAAAAAAAAFUYOpKNer1es2nTpmbKlClNq9Ua7X0CAAAAAAAAAACApmmapt/vN3v27GmOP/74pt3+5X8D7ogG4jZt2tTMnTv3V7JzAAAAAAAAAAAAELVhw4bmhBNO+KXbHNFA3JQpU5qmaZp2ux36C3G9Xu+It/25fr8fbjJ/tS6zzv82Xfg/ZV5/dI1BX6fU76aUQf0LiaXes6GhIzpl/DcjIyOjsCe/aJD3rZQSx+cgfz4BIGvChAnhJvqdWOq6OLNOqfuJQb7Xy6zT7XaLrJN5rymj1LGWUeJcUOK5SkmZ15NpDh06FG4AAAAAgP/y8zm2X+aIJkh+/iC11WqFHqoO8oPeEutk9ktT3z8qlFDqPavtGCj1+6zt9UQZogNg0JX4Dh3k69VBvu4Y5HUGuWFw1XZMD+IaAAAAAEDdjuQ54+D+11wAAAAAAAAAAAAIMBAHAAAAAAAAAABAFQzEAQAAAAAAAAAAUAUDcQAAAAAAAAAAAFTBQBwAAAAAAAAAAABVMBAHAAAAAAAAAABAFQzEAQAAAAAAAAAAUAUDcQAAAAAAAAAAAFTBQBwAAAAAAAAAAABVGIps3Ov1mlardcTb9/v98A5lmowS65R6Le12fK6x2+2Owp78opp+nyXXif5OI5/LV9MMDw+Hm8zxmWky+5Z5DzJK/X56vV64ib7Xmc9A5rWU+qwBQCmZ77bMNdEgr5NR4vqmaXKvp9PphJuMQb0uGtTr1aYp956Vus7P3L8PDYUe9zRNM7jPSUrdT2Tes1LPVgAAAACAGH8hDgAAAAAAAAAAgCoYiAMAAAAAAAAAAKAKBuIAAAAAAAAAAACogoE4AAAAAAAAAAAAqmAgDgAAAAAAAAAAgCoYiAMAAAAAAAAAAKAKBuIAAAAAAAAAAACogoE4AAAAAAAAAAAAqmAgDgAAAAAAAAAAgCoYiAMAAAAAAAAAAKAKBuIAAAAAAAAAAACogoE4AAAAAAAAAAAAqjAU2bjdbjetVuuIt+92u+Edyuj3++Em8jpezTpRpd6zdjs+C5l5/YP8u8m8BxmZfev1eqHtO51OeI1Sx1rm9+nckZPZt+ixBgy2s846K9zceOON4eYHP/hBuHnggQfCzd69e8MNZJS4JsiskbkmylwPDLLMNfvIyEiRdUpde5U41gb5GreUUp+dzDGQea8PHz4cbqL3laXuJUod06V+NwAAAADA6PMX4gAAAAAAAAAAAKiCgTgAAAAAAAAAAACqYCAOAAAAAAAAAACAKhiIAwAAAAAAAAAAoAoG4gAAAAAAAAAAAKiCgTgAAAAAAAAAAACqYCAOAAAAAAAAAACAKhiIAwAAAAAAAAAAoAoG4gAAAAAAAAAAAKiCgTgAAAAAAAAAAACqYCAOAAAAAAAAAACAKhiIAwAAAAAAAAAAoApDkY273W7oh7fb8Xm7fr8fbjqdTpF1er1eaPtSr782g/wetFqtcBM9bjKin82sUsd05n3OrJN5PSV+n00zuMcaUE7mHLVkyZJw82u/9mvhZt68eeFm8uTJ4ebuu+8ONzt27Ag3kPnezTQllLomKnXNXur1ZGSuvTL3rtF1BvlautQ1bql9GzduXLiZPXt2uLnkkkvCzXHHHRduDh8+HNp+37594TVWrlwZbh5//PFwM8ifAwAA6nPOOeeEm3e84x3hZtasWeEmep3fNE3z/PPPh7a///77w2v87Gc/CzcAABGe3AEAAAAAAAAAAFAFA3EAAAAAAAAAAABUwUAcAAAAAAAAAAAAVTAQBwAAAAAAAAAAQBUMxAEAAAAAAAAAAFAFA3EAAAAAAAAAAABUwUAcAAAAAAAAAAAAVTAQBwAAAAAAAAAAQBUMxAEAAAAAAAAAAFAFA3EAAAAAAAAAAABUwUAcAAAAAAAAAAAAVTAQBwAAAAAAAAAAQBWGRvOH93q9cNPpdMJNt9sNN61Wa9SbzOvPyLxn/X5/FPbkF2Xe53Y7PqeZea9L/X6GhuIfsxL7ljkGSjUZmeOm1Dqljs8S57WMUp81qE3mfDN27NhwkzlPX3bZZeGm1L5973vfC21/+PDh8BrUp8Q1Tonv3JIG+X4i05S6dx3UY63UNW6pa/aME044IdwsWrQo3Fx00UXhZsmSJeFmxowZ4ebJJ58Mbf/ss8+G13j88cfDTYZ7EAAAsubNmxdufud3fifcvPOd7ww3s2bNCjeZa+NXXnkltP19990XXmPp0qXh5u677w43AKVkztEf+MAHws0DDzwQbp566qlwMzw8HG5g0Azu02gAAAAAAAAAAAAIMBAHAAAAAAAAAABAFQzEAQAAAAAAAAAAUAUDcQAAAAAAAAAAAFTBQBwAAAAAAAAAAABVMBAHAAAAAAAAAABAFQzEAQAAAAAAAAAAUAUDcQAAAAAAAAAAAFTBQBwAAAAAAAAAAABVMBAHAAAAAAAAAABAFQzEAQAAAAAAAAAAUAUDcQAAAAAAAAAAAFRhaDR/eKvVCjfdbrfIOv1+P9yUMDQU/5W88Y1vDDdjx44NNy+99FK42bJlS7gZHh4ON5nfZ7sdnwfNrDMyMhJuojqdTrgZ1M9A0+Q+06XON7Xp9Xqh7TPvM5CT+f5YsWJFuHniiSfCzbRp08LNW97ylnCzdevWcPP888+Htn/00UfDa1CfzLVU9Ds0c92RuV4tdW9U6nqt1L6Veq+jx01mnVLvc6n7ifHjx4eb448/PtxcccUV4ebKK68MN6eeemq42bx5c7h58MEHw80Pf/jD0PYvvPBCeI1nnnkm3GQ+N5nzeuZZROYZDowbNy7cnHLKKeHmhBNOCDdjxowJN5nPaOY9yFxLHT58OLR95jO9a9eucLNv375wk7k/3L59e7jJPP8FqM21114bbpYsWRJuNm7cGG7uueeecDNhwoRws2jRotD2119/fXiNefPmhZuHH3443GS+dwEyzjvvvHDzla98JdwsXbo03Nx8883hJvNsaZDnD3h98hfiAAAAAAAAAAAAqIKBOAAAAAAAAAAAAKpgIA4AAAAAAAAAAIAqGIgDAAAAAAAAAACgCgbiAAAAAAAAAAAAqIKBOAAAAAAAAAAAAKpgIA4AAAAAAAAAAIAqGIgDAAAAAAAAAACgCgbiAAAAAAAAAAAAqIKBOAAAAAAAAAAAAKpgIA4AAAAAAAAAAIAqGIgDAAAAAAAAAACgCkOj+cP7/X64abVao7AnvxqdTie0fbfbDa8xZsyYcPPhD3843Jx66qnh5qGHHgo3999/f7h56qmnws2ePXvCzfDwcLjJHNPt9ujPnfZ6vXCT2a/MOhmZ80Bm3zLrZI6BQRZ9D0qd12t7n6GUH//4x+Hmu9/9briZNWtWuFm0aFG4ufTSS8PNhg0bQtuvWbMmvMbBgwfDDYMtc90+qN9Vme/dzOuP3hs1Te49K9WUuv7MqOl6bcKECeFmwYIF4ebKK68MN1dffXW4yXwfZr53vvWtb4Wb2267Ldzs378/tH2p42ZoKP7oKnNey6wzyM+wGFzHHHNMuPnUpz4Vbj7wgQ+Em/Hjx4ebUt+hu3fvDjfRfZs8eXJ4jcxzzC1btoSbzD3IT3/603Dz13/91+Fm+/bt4QaglMz31BlnnBFuMt9TX/rSl8LN97///XAzbdq0cPOhD30otP0NN9wQXuPMM88MNyMjI+EGoJTZs2cXWSczG/LCCy8UaaL/PgOjzV+IAwAAAAAAAAAAoAoG4gAAAAAAAAAAAKiCgTgAAAAAAAAAAACqYCAOAAAAAAAAAACAKhiIAwAAAAAAAAAAoAoG4gAAAAAAAAAAAKiCgTgAAAAAAAAAAACqYCAOAAAAAAAAAACAKhiIAwAAAAAAAAAAoAoG4gAAAAAAAAAAAKiCgTgAAAAAAAAAAACqYCAOAAAAAAAAAACAKgxFNm61Wk2r1Tri7fv9fniHIj//1azT6XTCTbfbDW2feS0TJkwIN+9///vDzZw5c8LN2972tnBzxRVXhJs77rgj3PzgBz8IN08++WS4yRxrGb1ebyDXKPX6M+u02/H53uhnOqvUea3E7yfzPpc4noH/K3MeuOWWW8LNlClTijRnnHFGuLngggtC25922mnhNR5//PFwQ32i3++lrgcySn2/Z9bJKHUdlfmdljgOMmtkTJw4MdwsXLgw3Hz2s58NNxdddFG4OXz4cLhZuXJluPna174Wbu69995wMzIyEm6ix06pc1SpYzqj1D0ldZk9e3a4mTx5crjZunVruNmyZUu4WbduXbjZvn17uFmwYEG4Of3000Pb79+/P7zGtGnTws2xxx5bpHnnO98ZbmbNmhVu/uAP/iDcZL53ATIy16yZ+92pU6eGm8y15PDwcLh5+eWXw80//uM/hrbPvGd/9Ed/FG78WwMwyNauXfta78L/08knnxxuMvc6GzZsCDcwmvyFOAAAAAAAAAAAAKpgIA4AAAAAAAAAAIAqGIgDAAAAAAAAAACgCgbiAAAAAAAAAAAAqIKBOAAAAAAAAAAAAKpgIA4AAAAAAAAAAIAqGIgDAAAAAAAAAACgCgbiAAAAAAAAAAAAqIKBOAAAAAAAAAAAAKpgIA4AAAAAAAAAAIAqGIgDAAAAAAAAAACgCkORjVutVtNqtY54+16vF96hTqcTbrrdbpEmqt/vh5t9+/aFmz/+4z8ON5///OfDzfHHHx9uLr744nBz3nnnhZsrrrgi3Hz2s58NN6tXrw43mc9BCZnjs92Oz9BGzhk/l/l8Dur73DS59zoj814DZM5RBw4cCDfDw8PhJmPGjBmh7d/0pjeF13j88cfDDYMtcx0R/d7NfE+X+m7PXHtl7ttKvM9NM9j7llkneg2euWafNm1auPnDP/zDcPO2t70t3MyePTvcZI7p5cuXh5svfelL4eb5558PN6XudUqccwb53qjU+QZ+8pOfhJvf/u3fDjeZ4zPzGR0ZGSnSDA2FHmM3TdM006dPD22fef1jxowJN6effnq4uf7668PNhz/84XAzf/78cANQm29961vh5vLLLw83f/InfxJuTjjhhHCTeT2bN28ObX/nnXeG1/jkJz8ZbsaOHRtuSj2TBHj00UfDzaZNm8JNZmbjuuuuCzf33XdfuHniiSfCDYwmfyEOAAAAAAAAAACAKhiIAwAAAAAAAAAAoAoG4gAAAAAAAAAAAKiCgTgAAAAAAAAAAACqYCAOAAAAAAAAAACAKhiIAwAAAAAAAAAAoAoG4gAAAAAAAAAAAKiCgTgAAAAAAAAAAACqYCAOAAAAAAAAAACAKhiIAwAAAAAAAAAAoAoG4gAAAAAAAAAAAKiCgTgAAAAAAAAAAACqMBTZuNfrhX54q9UKbZ9ZozaHDh0KN9/+9rfDzbp168LN1VdfHW7e/e53h5sFCxaEm8WLF4ebf/qnfwo3TzzxRLjJ2LJlS2j7xx57LLzGQw89FG7Gjx8fbt7xjneEmylTpoSbzO/zhRdeCDcZpc6F/X4/3HQ6ndD2r/dzNNTooosuCjcXX3xxuJkzZ064yZgwYUJo++OOOy68RvTc2TRN0+12ww3lZH6n0e/EzPd05hoi81oy65SS2bdS7/WgrjN9+vTwGh/84AfDzRVXXBFuZsyYEW7+/d//Pdzcfvvt4ebee+8NN+vXrw83IyMj4SZzDV7iWMt8BjIy67Tb8f//Wer1QMbhw4df6114zWXOn9u2bRuFPXn1zj333HCzcOHCcJN5/V/+8pfDjWdFQG0y9wZf+MIXws2nP/3pcPO5z30u3Bx11FHh5qabbgpt/+yzz4bXuOaaa8LNvn37wg3AIJs4cWKRdQ4cOBBu3IdSA38hDgAAAAAAAAAAgCoYiAMAAAAAAAAAAKAKBuIAAAAAAAAAAACogoE4AAAAAAAAAAAAqmAgDgAAAAAAAAAAgCoYiAMAAAAAAAAAAKAKBuIAAAAAAAAAAACogoE4AAAAAAAAAAAAqmAgDgAAAAAAAAAAgCoYiAMAAAAAAAAAAKAKBuIAAAAAAAAAAACogoE4AAAAAAAAAAAAqjAU2bjVajWtVuuIt+/1euEd6vf74SYj8jpKrpF5z3bv3h1uHnzwwXCza9eucPPyyy+Hm6uuuircnHvuueFmwYIF4WbOnDnhZsyYMeFm3759oe3f8573hNfYunVruMkc07NmzQo3Bw8eDDd79+4NNzfffHO4yezbIOt2u6/1LgCvsUsuuSTcvPWtbw03M2fODDeZ66ItW7aEtl+1alV4DefO+pS6B4nK7Nfw8HC4abfj/08qc12YeT2ZdUrdh2Vk1jnuuONC21977bXhNa6//vpwM23atHCzfPnycPPd73433Dz66KPh5pVXXgk3hw4dCjeZ75BOpxNuSsh8pjPnm4zMvpU6RwGD6+ijjw43733ve8NN5nv3lFNOCTePP/54uMl8h46MjIQbgEF2+PDhcPOd73wn3GS+d37jN34j3CxZsiTcPPLII6Ht77zzzvAaTz/9dLgBGGSZ5wqlnklOnjw53Jx22mnhJvPdlplZgCPlL8QBAAAAAAAAAABQBQNxAAAAAAAAAAAAVMFAHAAAAAAAAAAAAFUwEAcAAAAAAAAAAEAVDMQBAAAAAAAAAABQBQNxAAAAAAAAAAAAVMFAHAAAAAAAAAAAAFUwEAcAAAAAAAAAAEAVDMQBAAAAAAAAAABQBQNxAAAAAAAAAAAAVMFAHAAAAAAAAAAAAFUwEAcAAAAAAAAAAEAVhl7rHfif2u34jF6v1ws3/X4/3ET3LbNfmdefcfjw4XCzdu3acLNz585w8/TTT4ebhQsXhpsZM2aEm1NPPTXczJ8/P9zMnTs3tP0xxxwTXmPBggXhppRNmzaFm2nTpo3CnvyiTqcTbjLnm1arFW4y5xygLpMmTQo38+bNCzfTp08PNxn79+8PNy+++GJo+xUrVoTXgKYpc2+QuR4odT9RSqnrqFJOOeWUcPPud787tP11110XXiN6/9E0TXPrrbeGm2XLloWb1atXh5u9e/eGm5GRkXBT6lgr9Wwh+nkrdb7JvP5Mk7nXG+TzDZSSORfMnDkz3Fx22WXh5uSTTw5tf/zxx4fXyDyTHBqKP5JfunRpuLnrrrvCTeZZLgC5e5Dbbrst3GT+vemtb31ruLnkkktC2z/22GPhNV5++eVwAzDIxo0bF24mT548Cnvyq1Hi2RKMtrr+tQQAAAAAAAAAAIDXLQNxAAAAAAAAAAAAVMFAHAAAAAAAAAAAAFUwEAcAAAAAAAAAAEAVDMQBAAAAAAAAAABQBQNxAAAAAAAAAAAAVMFAHAAAAAAAAAAAAFUwEAcAAAAAAAAAAEAVDMQBAAAAAAAAAABQBQNxAAAAAAAAAAAAVMFAHAAAAAAAAAAAAFUwEAcAAAAAAAAAAEAVhiIb9/v9pt/vj9a+/OcaUa1Wq8g6vV4v3ESN9vv7ahw4cCDcPP/88+Fm3bp14eaHP/xhuJk9e3a4OeWUU8LN/Pnzw82pp54a2n769OnhNc4999xws2DBgnCTsXPnznDz4IMPhpvMZ7rb7YabUkqdC6Pa7fjsdYnzLQy6MWPGhJt3vOMd4WbRokXhZvLkyeEmI3NN8Mgjj4S23717d3gN6lPi+zDzPT2o3+1NU27fSq2TaTL3E5deemm4ue6660LbZ+5ZnnjiiXDz1a9+Ndw899xz4WZ4eDjcZK4/M8daRqlzQeZ6usR7UOrz2el0wg3QNNOmTQs3b3vb28JN5pnUu9/97nATfT1r1qwJr5F5HrV27doi66xfvz7cAFDOs88+G25WrFgRbi688MJwc/HFF4e2/7d/+7fwGnfddVe4AahN5t+BSsnMhuzZs2cU9gTy/IU4AAAAAAAAAAAAqmAgDgAAAAAAAAAAgCoYiAMAAAAAAAAAAKAKBuIAAAAAAAAAAACogoE4AAAAAAAAAAAAqmAgDgAAAAAAAAAAgCoYiAMAAAAAAAAAAKAKBuIAAAAAAAAAAACogoE4AAAAAAAAAAAAqmAgDgAAAAAAAAAAgCoYiAMAAAAAAAAAAKAKBuIAAAAAAAAAAACowlA0aLVao7Ltz/X7/SJNZt+iMvvV6XTCTea1DA8Ph5uMzL51u91ws3379iLNf/zHf4SbzHswc+bM0PbHHntseI3f+73fCzfz5s0LN7t27Qo3jz76aLhZtWpVuBkZGQk37XZ8jrjX64WbjMw5JypzPJd6/TDIxo8fH27OOeeccPPpT3863Jx99tnhZsyYMeHm+eefDzff//73w833vve9cAOZ7/eoUvc5pe6NBvleJ3PtMWnSpHDz5je/Ody85z3vCTfRa/B169aF17j11lvDzbPPPhtuMr+bUveUJZ4RNE2Z803TlLk3yPw+M68/05Q6R7nXoTZnnXVWuMk8X1q0aFG4ydxTPf3006HtV65cGV5j2bJl4SbzfBEAmqZpNm3aFG527NgRbk466aTQ9vPnzw+vcdddd4UbgEE2NBQevSn2PCpj3Lhx4WbChAnh5tChQ+EGjpS/EAcAAAAAAAAAAEAVDMQBAAAAAAAAAABQBQNxAAAAAAAAAAAAVMFAHAAAAAAAAAAAAFUwEAcAAAAAAAAAAEAVDMQBAAAAAAAAAABQBQNxAAAAAAAAAAAAVMFAHAAAAAAAAAAAAFUwEAcAAAAAAAAAAEAVDMQBAAAAAAAAAABQBQNxAAAAAAAAAAAAVMFAHAAAAAAAAAAAAFUYigb9fn9Utv25VqsVbjIy+1ZCr9cLN4P6WpqmaUZGRl7rXfiVarfLzJAODw+Htp87d254jeOOOy7cZKxZsybc/PM//3O42bFjR7jJfN4y56hOpxNuut1uuMnsW6lzbtQgn9dg6tSp4eaCCy4IN5/4xCfCzSWXXBJuMueoTZs2hZtvfOMb4eZrX/tauFm3bl24gcz3YYnvqsx+Za4hhobCt4XFrqMyMuucccYZ4eb9739/uLnwwgvDzfbt20Pb33fffeE1br/99nCTOQZK3e8O8rGWeT2ZJnPvGn09mfv9UueozHuW2bfMdRSUkjlHZe51Zs+eHW727dsXbrZu3RpuJk6cGNr+8ssvD6+xZ8+ecJN5/S+++GK48WwFoD5btmwJN6+88kq4mTdvXrgB6pK5zp8wYUK4OXjwYLgp8W+okyZNCq+R+XegUnbu3BlunnvuuSLrwGjyF+IAAAAAAAAAAACogoE4AAAAAAAAAAAAqmAgDgAAAAAAAAAAgCoYiAMAAAAAAAAAAKAKBuIAAAAAAAAAAACogoE4AAAAAAAAAAAAqmAgDgAAAAAAAAAAgCoYiAMAAAAAAAAAAKAKBuIAAAAAAAAAAACogoE4AAAAAAAAAAAAqmAgDgAAAAAAAAAAgCoYiAMAAAAAAAAAAKAKQ6/1DvxPrVYr3PT7/VHYk9dGr9cLN5n3LNNk1PS7aZrc7yfjsssuC23/+c9/PrzGwoULw81TTz0VbpYtWxZuli9fHm5KfXYyut1ukXUyn7dok3nP2u347HWp9wwmT54cbq688spwc9NNN4Wbk046KdxkzoUvvvhiuPnzP//zcLN06dJws23btnADGZnv0Oh3YmaNzPdh5nu3xOvPrpN5D44++uhwc91114Wbt771reHmwIED4eb+++8PbZ85R+/evTvclDpuBvl+N/O9m3nfMp/rEveumf3K/G5GRkbCTcbQUPwRWalnBJCROd889thj4eb3f//3w82ZZ54ZbjLnjyVLloS2P+ecc8JrnH766eHm1FNPDTdf/OIXw8369evDDQCDbdeuXeFm69at4WbKlCmh7d/ylreE13jkkUfCTeZaBWiaxYsXh5u/+7u/CzcnnnhiuNm3b1+42bBhQ7iJngszzwgyzyRLyTy/2LNnzyjsCZTlL8QBAAAAAAAAAABQBQNxAAAAAAAAAAAAVMFAHAAAAAAAAAAAAFUwEAcAAAAAAAAAAEAVDMQBAAAAAAAAAABQBQNxAAAAAAAAAAAAVMFAHAAAAAAAAAAAAFUwEAcAAAAAAAAAAEAVDMQBAAAAAAAAAABQBQNxAAAAAAAAAAAAVMFAHAAAAAAAAAAAAFUwEAcAAAAAAAAAAEAVhiIbt1qtptVqjda+NE3TNL1eb1R//s+12/FZwBL7lnl/+/3+636dTqcTbrrdbrjJyLwHH/nIR0Lbz58/P7xGZr+ee+65cPPQQw+Fm8xnrdSxVpvMuRAG1TnnnBNuPvGJT4SbD33oQ+Fm6tSp4SbjmWeeCTdf//rXw83tt98ebrZt2xZuoJTMNUH0eiVzvVqbUtcdixcvDjeLFi0KNzNnzgw3P/rRj8LNsmXLQtvv3r07vEbGaD8beDXrjB07NtwMDw+Hm0G+d8183qKvp9T9VOYYyDSlnkfBIHv55ZfDzXe+850iTea8NmPGjND25557bniNz3zmM+Hm7W9/e7h58sknw83NN98cboCmGTNmTLi57rrrws03vvGNcAOZ+739+/eHm+h36IIFC8JrjB8/PtwATXPxxReHmzvuuCPcTJ48OdxkZM4FmWdyr3eTJk0KN/4dnRqYhAAAAAAAAAAAAKAKBuIAAAAAAAAAAACogoE4AAAAAAAAAAAAqmAgDgAAAAAAAAAAgCoYiAMAAAAAAAAAAKAKBuIAAAAAAAAAAACogoE4AAAAAAAAAAAAqmAgDgAAAAAAAAAAgCoYiAMAAAAAAAAAAKAKBuIAAAAAAAAAAACogoE4AAAAAAAAAAAAqmAgDgAAAAAAAAAAgCoMjeYP7/f74abdjs/o9Xq9cJPZt1arNarbN03utZR6zzKvp9Qx0O12w03GmDFjws1ZZ50Vbk477bTQ9lOmTAmvsW/fvnCzefPmcLNly5ZwU+qYHmSdTifcZD4H0fet1HkNPvrRj4abG264IdxccMEF4WbixInhZu/eveHmrrvuCjd/+qd/Gm42bNgQbl555ZVwA693me/pzDVR5vq71DV7Zp3JkyeHm6uuuirczJs3L9xkrnMffvjhcPPTn/40tH2pa6/MdWHmGvdd73pXuHnve98bbvbv3x9uli9fHm7uvvvucFPqXjzalNqvzDoZg7xvMMgyn52MzLXUtm3bQtvff//94TXmz58fbm688cZw8853vjPcfPe73w03mesbGGQnnnhiuPnN3/zNcHPhhReGm3/5l38JNyMjI+GGulx66aXhZsmSJeFm+/btoe0fe+yx8BqPPPJIuAGaZseOHeEm83xtkGX+veXQoUOh7WfOnBleY5CNGzcu3GTmD2DQ+AtxAAAAAAAAAAAAVMFAHAAAAAAAAAAAAFUwEAcAAAAAAAAAAEAVDMQBAAAAAAAAAABQBQNxAAAAAAAAAAAAVMFAHAAAAAAAAAAAAFUwEAcAAAAAAAAAAEAVDMQBAAAAAAAAAABQBQNxAAAAAAAAAAAAVMFAHAAAAAAAAAAAAFUwEAcAAAAAAAAAAEAVDMQBAAAAAAAAAABQhaHR/OH9fj/ctNvxGb1WqxVuMvs2iGs0TdP0er1wk3nPSr3Ppd63jMzxee2114abY445JrR95nezatWqcHP33XeHmy1btoSbjEE+pjPrZD7XJQzy55Oc8ePHh5t3vetd4Wbx4sWh7S+77LLwGmeeeWa4GTt2bLjZuHFjuLntttvCzV/91V+Fm6eeeircdLvdcAO1KfH9PsjXHZlr3My1Suace8YZZ4Sb0047LdxMnjw53GSuje+9995ws2vXrtD2g3xdnGkyx8C5554bbjLHdOaaYPny5eGm0+mEmxLf75n3LPNaSp0LMwb1vg3IOXDgQLjZsGFDuNm6dWu4Oe6448LNySefHG5KPceDUubOnRturrnmmnDz+OOPh5uRkZFwQ12GhuL/RPvmN7853CxcuDDcPPPMM6Ht77zzzvAanklCzpo1a8LNxz/+8XCzYMGCcPPss8+Gm8yzleHh4XATfb70sY99LLzGOeecE25Kefnll8PNjh07RmFPoCx/IQ4AAAAAAAAAAIAqGIgDAAAAAAAAAACgCgbiAAAAAAAAAAAAqIKBOAAAAAAAAAAAAKpgIA4AAAAAAAAAAIAqGIgDAAAAAAAAAACgCgbiAAAAAAAAAAAAqIKBOAAAAAAAAAAAAKpgIA4AAAAAAAAAAIAqGIgDAAAAAAAAAACgCgbiAAAAAAAAAAAAqIKBOAAAAAAAAAAAAKowFNm43++Hfnir1Qptn1kj25RQ6vVnlHqfM+9BKRMmTAg3b3rTm8LNtddeG25mzJgR2n7//v3hNX70ox+FmxUrVoSb4eHhcJNR6rMzyMd0iXNOqdc/qOf1QRc9dzRN03zsYx8LN1dddVW4OfPMM0PbH3300eE1Muebhx9+ONwsXbo03DzwwAPhZu3ateGmlMy54Kyzzgo3c+fODTfHH398aPvx48eH11i/fn24efLJJ8PN2LFjw83PfvazcHPo0KFwU5t2e/T/n1Dmu63EfjVN0/R6vXCTOQ9kXs/ixYvDTeb78ODBg+FmzZo14ebpp58ON1Glfp+ljIyMFFln0qRJ4SZzni51L17i/JE51jJNp9Mpsk7mc5DZN6AuU6dODTeZ+92dO3eGm1LfoTDIMtd406dPDzdbtmwJNzBnzpxwM2/evHCTuc5dtWpVaPvMvx0B5dxyyy3h5thjjw0327ZtCzel/h35oYceCm3/4osvhte48cYbw83ll18ebjIyswTdbncU9gTK8hfiAAAAAAAAAAAAqIKBOAAAAAAAAAAAAKpgIA4AAAAAAAAAAIAqGIgDAAAAAAAAAACgCgbiAAAAAAAAAAAAqIKBOAAAAAAAAAAAAKpgIA4AAAAAAAAAAIAqGIgDAAAAAAAAAACgCgbiAAAAAAAAAAAAqIKBOAAAAAAAAAAAAKpgIA4AAAAAAAAAAIAqGIgDAAAAAAAAAACgCkORjVutVtNqtY54+16vF96hjMg+/Vy/3x+FPSm/Rlap96zUe5B5PdOnTw83V199dbiZN29euOl0OqHt77vvvvAamWbTpk3hJvO7KXXuKHV81nQuyLyWzDFAzsyZM8PNb/3Wb4WbEue1UjZv3hxudu7cGW7OOeeccHP22WeHm/Hjx4ebzO8m8x16xhlnhJs3vOEN4WbOnDmh7YeGQpfDTdM0zYYNG8LN2rVrw824cePCzcqVK8PN8uXLw826devCzSAr8f3Wbsf/L9Ig309lZN6Ds846K9xkzoVbtmwJN+vXrw83e/bsCTfR46DUd27mHDV79uxwM2vWrHCTeQ8yx0DmOiLzOcjInD+i58LMa8mcb7vdbrgpdc51rwOcdNJJ4eaYY44JN08//XS4+dnPfhZuoDaZ7+pDhw6FmylTpoSbQf13LXIyz8re8573hJuFCxeGm9WrV4eb2267LbT91q1bw2sA5WS+P1566aVR2JPXzv79+0PbL1u2bNTXaJqmOf/888PNjBkzws2xxx4bbjLP5GDQ+AtxAAAAAAAAAAAAVMFAHAAAAAAAAAAAAFUwEAcAAAAAAAAAAEAVDMQBAAAAAAAAAABQBQNxAAAAAAAAAAAAVMFAHAAAAAAAAAAAAFUwEAcAAAAAAAAAAEAVDMQBAAAAAAAAAABQBQNxAAAAAAAAAAAAVMFAHAAAAAAAAAAAAFUwEAcAAAAAAAAAAEAVDMQBAAAAAAAAAABQhaHR/OGtVms0f/yr0m7HZwH7/f6obp81yO9zqX2bOHFiuDn99NPDzXvf+95wM3bs2HCzcePG0PbLli0Lr7Fq1apwk1Hqc1DK6/31DA3FvzZGRkbCDTmZ882MGTPCTafTCTcl9Hq9cHP22WeHm1mzZoWb0047LdwcOnQo3GQ+o5MnTw43kyZNCjeZ15P5nUbX2b9/f3iN2bNnh5t58+aFm8yxdumll4abcePGhZuvf/3r4Wb79u3hZpCVuCbIrFHiPie7TubcMXfu3HAzYcKEcLN69epws3Xr1nCTuT+Kfu9mvqcz1xCZ89qSJUvCzfnnnx9udu3aFW7uu+++cLNy5cpwk/nsdLvdcFPiXjxz7sjsV6YpdV5jcGWeE2WuC6nLnDlzws2CBQvCTebe6IUXXgg3O3fuDDdQm8xzhUwzPDwcbkpd41DG1VdfHW4+9rGPhZvMffWtt94abu64445wA/B69+CDD4ab73//++Hm13/918PN+PHjw80JJ5wQbtyLM2g87QMAAAAAAAAAAKAKBuIAAAAAAAAAAACogoE4AAAAAAAAAAAAqmAgDgAAAAAAAAAAgCoYiAMAAAAAAAAAAKAKBuIAAAAAAAAAAACogoE4AAAAAAAAAAAAqmAgDgAAAAAAAAAAgCoYiAMAAAAAAAAAAKAKBuIAAAAAAAAAAACogoE4AAAAAAAAAAAAqmAgDgAAAAAAAAAAgCoMRTZutVpNq9U64u37/X54hzJKrRMVea9+rtPphJuRkZFwk9m3TJP53QwNhQ7Lpmma5pRTTgk3H/jAB8LNGWecEW663W64+dd//dfQ9g8//HB4ja1bt4abjFKfz1LH5yB/djJNux2bi86cbyjn0KFD4eb5558PN5ljeurUqaHtM8dar9cLNyeccEK4ecMb3hBuSn2md+3aFW727t0bbjZt2hRuNm7cGG5eeumlcBP9ftu8eXN4jRdffDHcvPnNbw43ixYtCjfHHHNMuDnttNPCzemnnx5uHnrooXBTk8w1YfR7umly58LMeT2zb9OnTw83c+fODTeZe6rMOWr37t3hJmPs2LGh7Y866qjwGvPnzw83ixcvDjfXXHNNuJk4cWK4WbZsWbj59re/HW7Wrl0bbjKfnVL3IFGlrm9K3Odk16GM6L1E0zTNFVdcEW6iz2Kapmn2798fbihn3Lhxoe2vvPLK8BqZa/bM/cTKlSvDDZB7vjRp0qRwk7kHGT9+fLjxvROXuY7IPMP5zGc+E24yzyXvuOOOIk3mOTPA613m31q+/OUvh5u3vOUt4WbBggXh5sILLww35513Xrh54IEHwo1nOBwpfyEOAAAAAAAAAACAKhiIAwAAAAAAAAAAoAoG4gAAAAAAAAAAAKiCgTgAAAAAAAAAAACqYCAOAAAAAAAAAACAKhiIAwAAAAAAAAAAoAoG4gAAAAAAAAAAAKiCgTgAAAAAAAAAAACqYCAOAAAAAAAAAACAKhiIAwAAAAAAAAAAoAoG4gAAAAAAAAAAAKiCgTgAAAAAAAAAAACqMBTZuNfrNa1Wa7T2pWmapun3++Ems08l1sns18jISLgp9foz2u34zOXs2bPDzVVXXRVubrzxxnBz+PDhcLN27dpwc8stt4S2f+qpp8JrZI6BzO9zUD+f2XVKNRmZ96DX6436GqVeP02zcePGcPOFL3wh3Fx99dXhZvHixaHt9+zZE15j69at4WbSpEnhZu7cueEmc/7MvJ6VK1eGm8x3yPLly8PN+vXrw01NvvnNb4abzHFz9NFHh5sDBw6Em927d4eb2kS/3zqdTniNzL1B5rgpdX2zffv2cLN3795wM23atHCzb9++cBO9jmqappk6dWq4mTVrVmj7973vfeE1Mt/tJ554Yrg5dOhQuLntttvCzbJly8JN5r5ttJ+PvBqDum+l7g0yn89S50/iMtfff/ZnfxZuNmzYEG5WrVoVbjLnQnLe/va3h7a/4YYbwmtkrr+/9rWvhZvvfOc74QZomh07doSbzPOLD37wg+Emc826evXq0PZjx44NrzE8PBxudu7cGW4OHjwYbjL3IJ/85CfDzUc+8pFwk/k++Id/+Idw8+UvfzncPPHEE+EGgDJ+8pOfhJsVK1aEm8x36BVXXBFuMs9lM/fvmeeFvD75C3EAAAAAAAAAAABUwUAcAAAAAAAAAAAAVTAQBwAAAAAAAAAAQBUMxAEAAAAAAAAAAFAFA3EAAAAAAAAAAABUwUAcAAAAAAAAAAAAVTAQBwAAAAAAAAAAQBUMxAEAAAAAAAAAAFAFA3EAAAAAAAAAAABUwUAcAAAAAAAAAAAAVTAQBwAAAAAAAAAAQBWGRvOH9/v90fzx/6nVag3kOr1eb5T25L8r9T5nZN6DN73pTeHmXe96V7jJ7Nvw8HC4+cIXvhBunnzyydD2pY61UuuUkjl3ZD5v7XZ89jjzXpc4Fwzy66dpDh48GG6WL18ebu65555wkzkOojLHTaYZM2ZMkebAgQPhJvMZzTQ+o2Vk3uctW7aMwp7Ur8R3Veaz1ul0wk2p7+putxtu9u7dG27+4i/+Itx87nOfCzcf/ehHw80b3/jGcLNt27Zwc9xxx4W2P/XUU8NrjBs3Ltzs2bMn3PzlX/5luLn99tvDzcaNG8NN5t6g1P1E5vNW4tors19DQ6P6GOo/Zd7nUscAcVOmTAk3J510Uri55pprws0LL7wQbl566aVwQ9O8/e1vDzc33XRTaPtjjjkmvMbNN98cbv7mb/4m3GSeSQK58/Stt94abj7+8Y8XWSd6LZW5zs88X8xcF44dOzbcZJ6TZJ6vXXbZZeFm06ZN4Wb79u3hBgDWrVsXbjLPZTPXERdddFG4ueGGG8LNI488Em7ci78++QtxAAAAAAAAAAAAVMFAHAAAAAAAAAAAAFUwEAcAAAAAAAAAAEAVDMQBAAAAAAAAAABQBQNxAAAAAAAAAAAAVMFAHAAAAAAAAAAAAFUwEAcAAAAAAAAAAEAVDMQBAAAAAAAAAABQBQNxAAAAAAAAAAAAVMFAHAAAAAAAAAAAAFUwEAcAAAAAAAAAAEAVDMQBAAAAAAAAAABQhaHXegd+FXq9XrhptVqjsCevXma/+v1+kXUyzdBQ/BCbP39+uFm0aFG46XQ64ebw4cPh5ic/+Um42b9/f2j7brcbXiOj1PGZUWqdjMy+tdvxeeXMubCEQd0v8oaHh1/rXXhNZc65Bw8eHIU9AX5VSlxPD/I1e2bfMk3mWvrOO+8MN+973/vCzXnnnRduzj777HCTMWbMmND2IyMj4TV+/OMfh5u//du/DTcrV64MNzt27Ag3mfeglMxnJ3PvWkJmvwb53iBz3GTu24jbtm1buHniiSfCzfXXXx9uZsyYEW5uvvnmcPPTn/403JRy9NFHh5vrrrsu3Hzyk58MN0cddVRo+8x329KlS8PN5s2bww2Qk7mWvOWWW8LN3XffHW6++tWvhpvzzz8/tH30XqJpmmb16tXhZvv27eEmc1348MMPh5t777033GSuIwCglPvvvz/cXHPNNeFm5syZ4SZj8uTJ4WbcuHGjsCfUyJM7AAAAAAAAAAAAqmAgDgAAAAAAAAAAgCoYiAMAAAAAAAAAAKAKBuIAAAAAAAAAAACogoE4AAAAAAAAAAAAqmAgDgAAAAAAAAAAgCoYiAMAAAAAAAAAAKAKBuIAAAAAAAAAAACogoE4AAAAAAAAAAAAqmAgDgAAAAAAAAAAgCoYiAMAAAAAAAAAAKAKBuIAAAAAAAAAAACowtBrvQP/U7sdn9FrtVrhptfrjXqT2a+MTqcTbrrdbrjJ/G4uuOCCcHP++eeHm7Fjx4abHTt2hJtly5aFmz179oSbzPEZlTk++/3+KOzJa2eQ34MSx0BG5j0rdY4GgJKi31WZa+mMzLVKpsncg2S+33fv3h1ubrrppnBz9NFHh5tSv9MxY8aEtj906FB4jVdeeSXcPPXUU+Fm37594SYj87vJHJ+Zz06p5wSZe/6ozPtc6n6q1PtMGS+99FK4+eIXvxhufvd3fzfcXH755eHm5JNPDjcbNmwIN5nvg6lTp4abOXPmhJsTTzwx3GSe/X3lK18Jbb906dLwGs8991y4AQZb5nn+mjVrws2nPvWpcHPccceFts9cr2X+3eTgwYPhZmRkJNxk7lsy1xEAMMjuueeecPOtb30r3Bx77LHhJvOMdfLkyUUaXp/8hTgAAAAAAAAAAACqYCAOAAAAAAAAAACAKhiIAwAAAAAAAAAAoAoG4gAAAAAAAAAAAKiCgTgAAAAAAAAAAACqYCAOAAAAAAAAAACAKhiIAwAAAAAAAAAAoAoG4gAAAAAAAAAAAKiCgTgAAAAAAAAAAACqYCAOAAAAAAAAAACAKhiIAwAAAAAAAAAAoAoG4gAAAAAAAAAAAKjC0Gj+8HY7Pm/X6/XCTavVCjeZfet2u6HtM/vV7/eLNJnXn7Fw4cJws2jRonAzPDwcbtasWRNu/v7v/z7cHD58ONxEf6eZY63U8Zk51kp9DkrJ7Fup3090ncx+Zc7rAFBSp9MJN9Hv3VL3OaUM8utZtWpVuMkcA9H7w6YZ3Hu3EvfH2XUyx9og37eUup4ucdxkjoGhofhjqFLnm0E+r73e7du3L9zccccdo7Anv+h973tfuLngggvCzZIlS8JN5jM6fvz4cLN58+Zws2LFinDzgx/8INzcc889oe3Xr18fXmOQn0cBg2316tVFGgCgLpnnF7fccku4eeaZZ8LNscceG27WrVsXbjZu3BhueH3yF+IAAAAAAAAAAACogoE4AAAAAAAAAAAAqmAgDgAAAAAAAAAAgCoYiAMAAAAAAAAAAKAKBuIAAAAAAAAAAACogoE4AAAAAAAAAAAAqmAgDgAAAAAAAAAAgCoYiAMAAAAAAAAAAKAKBuIAAAAAAAAAAACogoE4AAAAAAAAAAAAqmAgDgAAAAAAAAAAgCoYiAMAAAAAAAAAAKAKQ5GN2+1202q1jnj7brcb3qF2Oz6j1+v1ijTRfcusEXl/X806mfd54sSJ4Wb69Onhpt/vh5sVK1aEm1tvvTXcPPTQQ+EmI3oclDpuMjK/z1Iy+1bqHJXZt8xxEF0nswYADLqRkZFRX2OQv0M7nU64KXWNV+qeKnPvmnkPSrxvpV5/qXuQzDqDvG+l7g1KHGuZe6NSx1qpez0G17Zt28LN0qVLw83LL78cblauXBluZs6cGW6GhkKPfZumaZrDhw+Hm/Xr14ebH/3oR+Fm9erV4abUsy8AAIBBtnHjxnDzzW9+cxT2BMrytA8AAAAAAAAAAIAqGIgDAAAAAAAAAACgCgbiAAAAAAAAAAAAqIKBOAAAAAAAAAAAAKpgIA4AAAAAAAAAAIAqGIgDAAAAAAAAAACgCgbiAAAAAAAAAAAAqIKBOAAAAAAAAAAAAKpgIA4AAAAAAAAAAIAqGIgDAAAAAAAAAACgCgbiAAAAAAAAAAAAqIKBOAAAAAAAAAAAAKrQ6vf7/f9to927dzfTpk1r2u1202q1jviH93q98A4dwe78SkReR1ap15LR6XTCzVFHHRVuLr300nCzYMGCcPPMM8+Em7vvvjvc7N69O9xkPgfR43OQj7WMzOez3Y7P93a73XCT2bfM76fEOappyhw7pd4zAMiaNGlSuIl+V9V0n5OVeQ8y19KZ68JS12uZ1xNV6rp4aGgo3AzyMZD5fQ7qMdA0ZT4HpV5LxiDftx04cCDcAAAAAAD/ZdeuXc3UqVN/6Tb+QhwAAAAAAAAAAABVMBAHAAAAAAAAAABAFQzEAQAAAAAAAAAAUAUDcQAAAAAAAAAAAFTBQBwAAAAAAAAAAABVMBAHAAAAAAAAAABAFQzEAQAAAAAAAAAAUAUDcQAAAAAAAAAAAFTBQBwAAAAAAAAAAABVMBAHAAAAAAAAAABAFQzEAQAAAAAAAAAAUAUDcQAAAAAAAAAAAFSh1e/3+//bRrt3726mTZsW/uHtdnzertfrFVnnCF72q1ZijaYZ3NffNE3TarXCTeYYyKxTSon3OvP6S/1uBvk8kJHZt4xS71sJmWOt1DkKAJqmaSZOnBhuarqfKKXUNXvmOiqzb91uN9xkfqfR15NZY5Dvp0rdH9b2vpXYt9p+N6X27cCBA+EGAAAAAPgvu3btaqZOnfpLt/EX4gAAAAAAAAAAAKiCgTgAAAAAAAAAAACqYCAOAAAAAAAAAACAKhiIAwAAAAAAAAAAoAoG4gAAAAAAAAAAAKiCgTgAAAAAAAAAAACqYCAOAAAAAAAAAACAKhiIAwAAAAAAAAAAoAoG4gAAAAAAAAAAAKiCgTgAAAAAAAAAAACqYCAOAAAAAAAAAACAKhiIAwAAAAAAAAAAoApDo/nD+/1+uGm1WgO7Tq/XCzdRnU4n3GT2q9R7llHqGChlUF9Pux2fh80cayU+NyXXych8rrvd7ijsyS/KHAclDPJnGgCaJv4dmrlWyXxPD/I9WKYpda2S+f2Uuj8qYZCPm0G+Dx3kz06JZx6l7nNKHQPuQQAAAACgHoM5CQEAAAAAAAAAAABBBuIAAAAAAAAAAACogoE4AAAAAAAAAAAAqmAgDgAAAAAAAAAAgCoYiAMAAAAAAAAAAKAKBuIAAAAAAAAAAACogoE4AAAAAAAAAAAAqmAgDgAAAAAAAAAAgCoYiAMAAAAAAAAAAKAKBuIAAAAAAAAAAACogoE4AAAAAAAAAAAAqjB0JBv1+/3UD892r2eZ96zU+1zbOqWUeD2DfNww2O/1IO8bAJQyqNdrta0zyNesNa0zyNd3g7xvpQzyZyfaDPJnGgAAAACo15E8Zzyigbg9e/a86p0ZNIP6ELbX673Wu/ArNajvM03T7XZf61143Rjkz7XPKAA0zYEDB17rXQAAAAAAAOAI7Nmzp5k2bdov3abVP4JpiF6v12zatKmZMmVK02q1fmU7CAAAAAAAAAAAAL9Mv99v9uzZ0xx//PFNu93+pdse0UAcAAAAAAAAAAAADLpfPi4HAAAAAAAAAAAA/58wEAcAAAAAAAAAAEAVDMQBAAAAAAAAAABQBQNxAAAAAAAAAAAAVMFAHAAAAAAAAAAAAFUwEAcAAAAAAAAAAEAVDMQBAAAAAAAAAABQhf8D/Oz2IBnQ998AAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 3200x3200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "x,_ = diffusion.sample(model, n=6,image_channels=args.image_channels)\n",
    "plot_images(x)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Denoise Images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "999it [00:31, 31.76it/s]\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 3200x3200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "torch.Size([10, 1, 32, 32])"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "set_seed(random_seed)\n",
    "denoise_img = diffusion.revert(model, n=1,image_channels=args.image_channels)\n",
    "plot_images(denoise_img)\n",
    "denoise_img.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Inspect Filters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x300 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x300 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "if f_settings is not None:\n",
    "    filter_size=f_settings['kernel_size']\n",
    "    beta=f_settings['kaiser_beta']\n",
    "\n",
    "    filters=[]\n",
    "    filters.append(circularLowpassKernel(f_settings['omega_c_down'], filter_size))\n",
    "    filters.append(circularLowpassKernel(f_settings['omega_c_up'], filter_size,beta=beta))\n",
    "\n",
    "    for filter in filters:  \n",
    "        plot_filter_and_response(filter)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Generate Rotational Images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "thatas=np.linspace(-90,90,9)\n",
    "# model paths\n",
    "modelpath_ref=\"E:\\Research\\Diffusion-Colab\\Results\\MNIST\\MNIST b16b0\\models_MNIST\\DDPM_Uncondtional_MNIST_0\\ckpt_MNIST_0.pt\"\n",
    "modelpath_test=\"E:\\Research\\Diffusion-Colab\\Results\\MNIST\\MNIST b16b2N\\models\\DDPM_Uncondtional_MNIST_3\\ckpt_MNIST_3.pt\"\n",
    "\n",
    "#modelpath_ref=\"E:\\Research\\Diffusion-Colab\\Results\\MNIST-M\\MNIST-M b16b0\\models\\DDPM_Uncondtional_MNIST-M_0\\ckpt_MNIST-M_0.pt\"\n",
    "#modelpath_test=\"E:\\Research\\Diffusion-Colab\\Results\\MNIST-M\\MNIST-M b16b2N\\models_mnistm_mnistm_b16b2N\\DDPM_Uncondtional_MNIST-M_3\\ckpt_MNIST-M_3.pt\"\n",
    "\n",
    "#modelpath_ref=\"E:\\Research\\Diffusion-Colab\\Results\\CIFAR\\CIFAR b16b0\\models_CIFAR\\DDPM_Uncondtional_CIFAR_0\\ckpt_CIFAR_0.pt\"\n",
    "#modelpath_test=\"E:\\Research\\Diffusion-Colab\\Results\\CIFAR\\CIFAR b16b1N\\models\\DDPM_Uncondtional_CIFAR_3\\ckpt_CIFAR_3.pt\"\n",
    "\n",
    "#name of the file to save results\n",
    "fname='MNISTM_rotational_results_arxiv.pkl'\n",
    "\n",
    "# for making videos\n",
    "fps=15\n",
    "ref_filename='video_MNIST_b16b0_0'\n",
    "test_filename='video_MNIST_b16b0_1'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# reference model\n",
    "f_settings_ref=None\n",
    "model_data_ref={\n",
    "    'unet_v': 0,\n",
    "    'args': args,\n",
    "    'f_settings': f_settings_ref,\n",
    "    'seed': random_seed,\n",
    "   'modelpath': modelpath_ref\n",
    "}\n",
    "\n",
    "# test model\n",
    "f_settings={}\n",
    "f_settings['kernel_size']=3\n",
    "f_settings['kaiser_beta']=2\n",
    "f_settings['omega_c_down'] =np.pi/2\n",
    "f_settings['omega_c_up'] = np.pi/2\n",
    "model_data_test={\n",
    "    'unet_v': 3,\n",
    "    'args': args,\n",
    "    'f_settings': f_settings,\n",
    "    'seed': random_seed,\n",
    "   'modelpath': modelpath_test\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Variant 0 Original UNet\n",
      "Unconditional UNet\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "999it [00:19, 50.91it/s]:00<?, ?it/s]\n",
      "999it [00:19, 51.21it/s]:19<02:37, 19.63s/it]\n",
      "999it [00:19, 51.54it/s]:39<02:16, 19.56s/it]\n",
      "999it [00:19, 51.35it/s]:58<01:56, 19.48s/it]\n",
      "999it [00:19, 50.93it/s]:17<01:37, 19.47s/it]\n",
      "999it [00:19, 51.37it/s]:37<01:18, 19.53s/it]\n",
      "999it [00:19, 51.90it/s]:57<00:58, 19.50s/it]\n",
      "999it [00:19, 50.30it/s]:16<00:38, 19.42s/it]\n",
      "999it [00:19, 50.97it/s]:36<00:19, 19.56s/it]\n",
      "100%|██████████| 9/9 [02:55<00:00, 19.53s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Variant 3 Modified UNet: filters around gelu + filters in up or downsampling\n",
      "Unconditional UNet\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "999it [00:31, 31.56it/s]:00<?, ?it/s]\n",
      "999it [00:32, 31.00it/s]:31<04:13, 31.65s/it]\n",
      "999it [00:32, 30.88it/s]:03<03:43, 32.00s/it]\n",
      "999it [00:33, 29.87it/s]:36<03:12, 32.16s/it]\n",
      "999it [00:32, 30.30it/s]:09<02:43, 32.67s/it]\n",
      "999it [00:31, 31.27it/s]:42<02:11, 32.78s/it]\n",
      "999it [00:32, 31.14it/s]:14<01:37, 32.50s/it]\n",
      "999it [00:31, 31.41it/s]:46<01:04, 32.36s/it]\n",
      "999it [00:31, 31.26it/s]:18<00:32, 32.19s/it]\n",
      "100%|██████████| 9/9 [04:50<00:00, 32.28s/it]\n"
     ]
    }
   ],
   "source": [
    "x_ref, results_ref =rotation_results(model_data_ref,thatas)\n",
    "x_test, results_test =rotation_results(model_data_test,thatas)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "fdata={\n",
    "    'x_ref': x_ref,\n",
    "    'x_test': x_test,\n",
    "    'results_ref': results_ref,\n",
    "    'results_test': results_test,\n",
    "    'modeldata_ref': model_data_ref,\n",
    "    'modeldata_test': model_data_test,\n",
    "    'thetas': thatas\n",
    "}\n",
    "\n",
    "with open(fname,'wb') as file:\n",
    "    pickle.dump(fdata,file)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[-90.  -67.5 -45.  -22.5   0.   22.5  45.   67.5  90. ]\n"
     ]
    }
   ],
   "source": [
    "# select an index from 0 to 3 \n",
    "im_i=2\n",
    "print(thatas)\n",
    "images_ref = torch.stack([tensor[im_i] for tensor in x_ref])\n",
    "images_test = torch.stack([tensor[im_i] for tensor in x_test])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 3200x3200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 3200x3200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_images(images_ref)\n",
    "plot_images(images_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "make_video(images_ref,ref_filename,fps=fps)\n",
    "make_video(images_test,test_filename,fps=fps)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "save_gif(ref_filename,fps,'128')\n",
    "save_gif(test_filename,fps,'128')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Inspecting rotational sampling across time steps"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 3200x3200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 3200x3200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "torch.Size([1, 32, 32])"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plot_images(results_ref[4][2::4])\n",
    "plot_images(results_test[7][2::4])\n",
    "results_ref[0][0].shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Misc. functionalities\n",
    "\n",
    "### Save MNIST dataset from CSV to PNG images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "from modules import utils\n",
    "\n",
    "current_directory = os.getcwd() \n",
    "csv_path = os.path.join(current_directory,\"data/MNIST/mnist_train_small.csv\")\n",
    "save_path= os.path.join(current_directory,\"trdata/MNIST\")\n",
    "\n",
    "utils.save_dataset_MNIST_CSV2PNG(csv_path,save_path)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Make Collage of Training Images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from modules.utils import make_collage\n",
    "\n",
    "gen_savepath=\"E:\\MNIST-M-6000\"\n",
    "\n",
    "image_size=32\n",
    "images_per_collage=400\n",
    "total_image_collage=2000\n",
    "make_collage(gen_savepath,gen_savepath,images_per_collage,total_image_collage,image_size)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "mydiffusion",
   "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": "3.9.18"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}