style_transfer.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "ControlLDM 最大的，\n",
    "ControlLDM.apply_model() return eps given x_noisy, t ,cond,\n",
    "cond = [c_concat] =IMAGE\n",
    "[c_crossattn] = text\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import open_clip\n",
    "from PIL import Image\n",
    "import torch\n",
    "model_clip, _, preprocess = open_clip.create_model_and_transforms('ViT-H-14', pretrained='../autodl-tmp/open_clip_pytorch_model.bin')\n",
    "model_clip = model_clip.to('cuda')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from torchvision.transforms import Resize \n",
    " \n",
    "torch_resize = Resize([224,224])\n",
    "im1_resize = torch_resize(img1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "tensor = torch.rand([1,3,2,2])\n",
    "gray_tensor = torch.sum(tensor * torch.tensor([0.299, 0.587, 0.114]).view(1, 3, 1, 1), dim=1, keepdim=True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "image = preprocess(Image.open(\"./test_imgs/building.png\")).unsqueeze(0)\n",
    "model_clip.encode_image(im1_resize.to('cuda'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "out_image = first_stage_model.encoder(image.to('cuda')).to('cuda')\n",
    "x.requires_grad_(True)\n",
    "optimizer = torch.optim.Adam([x], lr=0.1)\n",
    "num_iters = 200\n",
    "for cur_iter in range(num_iters):\n",
    "    print(cur_iter)\n",
    "    x_out = first_stage_model.encoder(x)\n",
    "    loss1 = (x_out - out_image).pow(2).mean()\n",
    "    loss1.backward(retain_graph=True)\n",
    "    optimizer.step()\n",
    "    optimizer.zero_grad(set_to_none=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "source1 = cv2.imread(\"./4.png\")\n",
    "input_image1 = cv2.cvtColor(source1, cv2.COLOR_BGR2RGB)\n",
    "input_image1 = HWC3(input_image1)\n",
    "img1 = resize_image(input_image1, 224)\n",
    "img1 = torch.tensor(np.array(img1)).permute(2,0,1).unsqueeze(0).cuda()\n",
    "img1 = img1.float() / 127.5 - 1.0\n",
    "model_clip.encode_image(img1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import cv2\n",
    "import einops\n",
    "import gradio as gr\n",
    "import numpy as np\n",
    "import torch\n",
    "import random\n",
    "\n",
    "from pytorch_lightning import seed_everything\n",
    "from annotator.util import resize_image, HWC3\n",
    "from annotator.canny import CannyDetector\n",
    "from cldm.model import create_model, load_state_dict\n",
    "from cldm.ddim_hacked import DDIMSampler\n",
    "from annotator.midas import MidasDetector\n",
    "from annotator.zoe import ZoeDetector\n",
    "\n",
    "preprocessor = None\n",
    "\n",
    "model_name = 'control_v11p_sd15_canny'\n",
    "model = create_model(f'./models/{model_name}.yaml').cpu()\n",
    "model.load_state_dict(load_state_dict('../autodl-tmp/v1-5-pruned.ckpt', location='cuda'), strict=False)\n",
    "model.load_state_dict(load_state_dict(f'./models/{model_name}.pth', location='cuda'), strict=False)\n",
    "model = model.cuda()\n",
    "ddim_sampler = DDIMSampler(model)\n",
    "first_stage_model = model.get_first_stage_model()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/root/miniconda3/envs/ldm/lib/python3.8/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
      "  from .autonotebook import tqdm as notebook_tqdm\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "No module 'xformers'. Proceeding without it.\n",
      "ControlLDM: Running in eps-prediction mode\n",
      "DiffusionWrapper has 859.52 M params.\n",
      "making attention of type 'vanilla' with 512 in_channels\n",
      "Working with z of shape (1, 4, 32, 32) = 4096 dimensions.\n",
      "making attention of type 'vanilla' with 512 in_channels\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "`text_config_dict` is provided which will be used to initialize `CLIPTextConfig`. The value `text_config[\"id2label\"]` will be overriden.\n",
      "`text_config_dict` is provided which will be used to initialize `CLIPTextConfig`. The value `text_config[\"bos_token_id\"]` will be overriden.\n",
      "`text_config_dict` is provided which will be used to initialize `CLIPTextConfig`. The value `text_config[\"eos_token_id\"]` will be overriden.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loaded model config from [./models/control_v11f1p_sd15_depth.yaml]\n",
      "Loaded state_dict from [../autodl-tmp/v1-5-pruned.ckpt]\n",
      "Loaded state_dict from [../autodl-tmp/control_v11f1p_sd15_depth.pth]\n"
     ]
    }
   ],
   "source": [
    "import cv2\n",
    "import einops\n",
    "import gradio as gr\n",
    "import numpy as np\n",
    "import torch\n",
    "import random\n",
    "\n",
    "from pytorch_lightning import seed_everything\n",
    "from annotator.util import resize_image, HWC3\n",
    "from annotator.canny import CannyDetector\n",
    "from cldm.model import create_model, load_state_dict\n",
    "from cldm.ddim_hacked import DDIMSampler\n",
    "from annotator.midas import MidasDetector\n",
    "from annotator.zoe import ZoeDetector\n",
    "model_name = 'control_v11f1p_sd15_depth'\n",
    "model_depth = create_model(f'./models/{model_name}.yaml').cpu()\n",
    "model_depth.load_state_dict(load_state_dict('../autodl-tmp/v1-5-pruned.ckpt', location='cuda'), strict=False)\n",
    "model_depth.load_state_dict(load_state_dict(f'../autodl-tmp/{model_name}.pth', location='cuda'), strict=False)\n",
    "model_depth = model_depth.cuda()\n",
    "ddim_sampler_depth = DDIMSampler(model_depth)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torchvision\n",
    "torchvision.utils.save_image(((one+1)*127.5)/255,\"3.png\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "one = Return_detected_map(path = \"./test_imgs/building.png\",detect_resolution = 512,low_threshold = 100,high_threshold=200,det = 'canny')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import cv2\n",
    "from annotator.util import HWC3\n",
    "import einops\n",
    "from PIL import Image\n",
    "from annotator.midas import MidasDetector\n",
    "from annotator.zoe import ZoeDetector\n",
    "import numpy as np\n",
    "def slerp(p0, p1, fract_mixing: float):\n",
    "    r\"\"\" Copied from lunarring/latentblending\n",
    "    Helper function to correctly mix two random variables using spherical interpolation.\n",
    "    The function will always cast up to float64 for sake of extra 4.\n",
    "    Args:\n",
    "        p0: \n",
    "            First tensor for interpolation\n",
    "        p1: \n",
    "            Second tensor for interpolation\n",
    "        fract_mixing: float \n",
    "            Mixing coefficient of interval [0, 1]. \n",
    "            0 will return in p0\n",
    "            1 will return in p1\n",
    "            0.x will return a mix between both preserving angular velocity.\n",
    "    \"\"\" \n",
    "    if p0.dtype == torch.float16:\n",
    "        recast_to = 'fp16'\n",
    "    else:\n",
    "        recast_to = 'fp32'\n",
    "    \n",
    "    p0 = p0.double()\n",
    "    p1 = p1.double()\n",
    "    norm = torch.linalg.norm(p0) * torch.linalg.norm(p1)\n",
    "    epsilon = 1e-7\n",
    "    dot = torch.sum(p0 * p1) / norm\n",
    "    dot = dot.clamp(-1+epsilon, 1-epsilon)\n",
    "    \n",
    "    theta_0 = torch.arccos(dot)\n",
    "    sin_theta_0 = torch.sin(theta_0)\n",
    "    theta_t = theta_0 * fract_mixing\n",
    "    s0 = torch.sin(theta_0 - theta_t) / sin_theta_0\n",
    "    s1 = torch.sin(theta_t) / sin_theta_0\n",
    "    interp = p0*s0 + p1*s1\n",
    "    \n",
    "    if recast_to == 'fp16':\n",
    "        interp = interp.half()\n",
    "    elif recast_to == 'fp32':\n",
    "        interp = interp.float()\n",
    "        \n",
    "    return interp\n",
    "def resize_image_to_squra(path, resolution):\n",
    "        \n",
    "    source = cv2.imread(path)\n",
    "    source = cv2.cvtColor(source, cv2.COLOR_BGR2RGB)\n",
    "    input_image = HWC3(source)\n",
    "    H, W, C = input_image.shape\n",
    "    H = float(H)\n",
    "    W = float(W)\n",
    "    k = float(resolution) / min(H, W)\n",
    "    H = resolution\n",
    "    W = resolution\n",
    "    H = int(np.round(H / 64.0)) * 64\n",
    "    W = int(np.round(W / 64.0)) * 64\n",
    "    img = cv2.resize(input_image, (W, H), interpolation=cv2.INTER_LANCZOS4 if k > 1 else cv2.INTER_AREA)\n",
    "    return img\n",
    "def Return_detected_map(path = \"./test_imgs/building.png\",detect_resolution = 512,low_threshold = 100,high_threshold=200,det = 'canny'):\n",
    "    input_image = cv2.imread(path)\n",
    "    input_image = cv2.cvtColor(input_image, cv2.COLOR_BGR2RGB)\n",
    "    if det == 'canny':\n",
    "        preprocessor = CannyDetector()\n",
    "    if det == 'Depth_Zoe':\n",
    "        preprocessor = ZoeDetector()\n",
    "    if det == 'Depth_Mias':\n",
    "        preprocessor = CannyDetector()\n",
    "    with torch.no_grad():\n",
    "        input_image = HWC3(input_image)\n",
    "        if det == 'canny':\n",
    "            preprocessor = CannyDetector()\n",
    "            detected_map = preprocessor(resize_image(input_image, detect_resolution), low_threshold, high_threshold)\n",
    "        if det == 'Depth_Zoe':\n",
    "            preprocessor = ZoeDetector()\n",
    "            detected_map = preprocessor(resize_image(input_image,detect_resolution ))\n",
    "        if det == 'Depth_Mias':\n",
    "            preprocessor = CannyDetector()\n",
    "\n",
    "        detected_map = HWC3(detected_map)\n",
    "        img = resize_image(input_image, detect_resolution)\n",
    "        H, W, C = img.shape\n",
    "        detected_map = cv2.resize(detected_map, (W, H), interpolation=cv2.INTER_LINEAR)\n",
    "        detected_map = torch.tensor(np.array(detected_map)).permute(2,0,1).unsqueeze(0).cuda()\n",
    "        detected_map = detected_map.float() / 127.5 - 1.0\n",
    "    return detected_map\n",
    "def learn_conditioning( model, ddim_sampler,img1, cond_base, uncond_base, ddim_steps, guide_scale, num_iters=200, cond_lr=1e-4):\n",
    "    cond = {\"c_crossattn\": [cond_base], 'c_concat': None}\n",
    "    un_cond = {\"c_crossattn\": [uncond_base], 'c_concat': None}\n",
    "    ldm = model\n",
    "    uncond_base.requires_grad_(True)\n",
    "    cond1 = cond_base.clone()\n",
    "    cond1.requires_grad_(True)\n",
    "    optimizer = torch.optim.Adam([cond1, uncond_base], lr=cond_lr) #\n",
    "    T = ddim_steps\n",
    "    ddim_sampler.make_schedule(T, verbose=False)\n",
    "    for cur_iter in range(num_iters):\n",
    "        L1 = ldm.get_first_stage_encoding(ldm.encode_first_stage(img1))\n",
    "        with torch.autocast('cuda'):\n",
    "            u = np.random.randint(T//3, 2*T//3)\n",
    "            t_u = ddim_sampler.ddim_timesteps[u]\n",
    "            tu = torch.tensor([t_u], device='cuda', dtype=torch.long)\n",
    "\n",
    "            cond[\"c_crossattn\"] = [cond1]\n",
    "            noise = torch.randn_like(L1)\n",
    "            x_t_u = ldm.sqrt_alphas_cumprod[t_u] * L1 + \\\n",
    "                ldm.sqrt_one_minus_alphas_cumprod[t_u] * noise\n",
    "            eps = ddim_sampler.pred_eps(x_t_u, cond, tu, unconditional_guidance_scale=guide_scale, unconditional_conditioning=un_cond)\n",
    "            loss1 = (eps - noise).pow(2).mean()\n",
    "            loss1.backward()\n",
    "            optimizer.step()\n",
    "            optimizer.zero_grad(set_to_none=True)\n",
    "\n",
    "    cond1.requires_grad_(False)\n",
    "    uncond_base.requires_grad_(False)\n",
    "    return cond1,uncond_base\n",
    "def Combine_condition(img1,img2, ldm, alpha = 0.5):\n",
    "    img1 = torch.tensor(np.array(img1)).permute(2,0,1).unsqueeze(0).cuda()\n",
    "    img2 = torch.tensor(np.array(img2)).permute(2,0,1).unsqueeze(0).cuda()\n",
    "    L1 = ldm.get_first_stage_encoding(ldm.encode_first_stage(img1.float() / 127.5 - 1.0))\n",
    "    L2 = ldm.get_first_stage_encoding(ldm.encode_first_stage(img2.float() / 127.5 - 1.0))\n",
    "    return slerp(L1,L2, alpha)\n",
    "def extract_into_tensor(a, t, x_shape):\n",
    "    b, *_ = t.shape\n",
    "    out = a.gather(-1, t)\n",
    "    return out.reshape(b, *((1,) * (len(x_shape) - 1)))\n",
    "def resize_image(input_image, resolution):\n",
    "    H, W, C = input_image.shape\n",
    "    H = float(H)\n",
    "    W = float(W)\n",
    "    k = float(resolution) / min(H, W)\n",
    "    H = resolution\n",
    "    W = resolution\n",
    "    H = int(np.round(H / 64.0)) * 64\n",
    "    W = int(np.round(W / 64.0)) * 64\n",
    "    img = cv2.resize(input_image, (W, H), interpolation=cv2.INTER_LANCZOS4 if k > 1 else cv2.INTER_AREA)\n",
    "    return img\n",
    "def tensor_to_array(input_image,time_num = 127.5):\n",
    "    if len(input_image.shape) == 3:\n",
    "        out  = einops.rearrange(input_image, 'c h w -> h w c')\n",
    "        if time_num == 127.5:\n",
    "             out = (out* 127.5 + 127.5).cpu().numpy().clip(0, 255).astype(np.uint8)\n",
    "        else:\n",
    "            out = (out* 255).cpu().numpy().clip(0, 255).astype(np.uint8)\n",
    "    else:\n",
    "        out  = einops.rearrange(input_image, 'b c h w -> b h w c')\n",
    "        if time_num == 127.5:\n",
    "             out = (out* 127.5 + 127.5).cpu().numpy().clip(0, 255).astype(np.uint8)\n",
    "        else:\n",
    "            out = (out* 255).cpu().numpy().clip(0, 255).astype(np.uint8)\n",
    "        out = out.squeeze(0)\n",
    "    return out\n",
    "\n",
    "source = cv2.imread(\"./b.png\")\n",
    "source = cv2.cvtColor(source, cv2.COLOR_BGR2RGB)\n",
    "prompt = 'best qualit, reality'\n",
    "a_prompt = '4k'\n",
    "n_prompt = 'longbody, lowres, bad anatomy, bad hands, missing fingers, pubic hair,extra digit, fewer digits, cropped, worst quality, low quality'\n",
    "num_samples = 1\n",
    "image_resolution = 512\n",
    "detect_resolution=512\n",
    "ddim_steps = 25\n",
    "guess_mode =None\n",
    "strength = 1\n",
    "scale = 5 \n",
    "seed = 4\n",
    "eta = 1\n",
    "low_threshold = 100\n",
    "high_threshold = 200\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tqdm\n",
    "@torch.no_grad()\n",
    "def invert(input_image,  model, image_resolution,\n",
    "           num_samples, a_prompt, n_prompt, prompt, ddim_sampler, control = None,guide_scale=7.5, num_inference_steps=100,\n",
    "           device='cuda',eta=1,get_image = False):\n",
    "  \n",
    "    # Encode prompt\n",
    "    input_image = HWC3(input_image)\n",
    "    img1 = resize_image(input_image, image_resolution)\n",
    "    H,W,C = img1.shape\n",
    "    ddim_sampler.make_schedule(num_inference_steps,verbose = False)\n",
    "    \n",
    "    intermediate_latents = []\n",
    "    timesteps = ddim_sampler_depth.ddim_timesteps\n",
    "    cond = { \"c_crossattn\": [model.get_learned_conditioning([prompt + ', ' + a_prompt] * num_samples)],\"c_concat\": [control] if control is not None else None}\n",
    "    un_cond = {\"c_crossattn\": [model.get_learned_conditioning([n_prompt] * num_samples)],\"c_concat\": None }\n",
    "    model.control_scales = ([1] * 13)\n",
    "\n",
    "    img1 = torch.tensor(np.array(input_image)).permute(2,0,1).unsqueeze(0).cuda()\n",
    "    L1 = model.get_first_stage_encoding(model.encode_first_stage(img1.float() / 127.5 - 1.0))\n",
    "    latents = L1\n",
    "    # Reversed timesteps <<<<<<<<<<<<<<<<<<<<\n",
    "    out = []\n",
    "    for i in range(1, num_inference_steps):\n",
    "\n",
    "        # We'll skip the final iteration\n",
    "        if i >= num_inference_steps - 1: continue\n",
    "\n",
    "        t = timesteps[i]\n",
    "        t = torch.tensor([t], dtype=torch.long, device='cuda')\n",
    "        \n",
    "        # Expand the latents if we are doing classifier free guidance\n",
    "        latent_model_input = latents\n",
    "\n",
    "        # Predict the noise residual\n",
    "        noise_pred =  ddim_sampler.pred_eps(latents, cond, t, unconditional_guidance_scale=guide_scale, unconditional_conditioning=un_cond)\n",
    "\n",
    "\n",
    "        current_t = max(0, t.item() - (1000//num_inference_steps)) \n",
    "        next_t = t \n",
    "        alpha_t = ddim_sampler.alphas_cumprod[current_t]\n",
    "        alpha_t_next = ddim_sampler.alphas_cumprod[next_t]\n",
    "        # Inverted update step (re-arranging the update step to get x(t) (new latents) as a function of x(t-1) (current latents)\n",
    "        #latents = (latents - (1-alpha_t).sqrt()*noise_pred)*(alpha_t_next.sqrt()/alpha_t.sqrt()) + (1-alpha_t_next).sqrt()*noise_pred\n",
    "        latents = latents*(alpha_t_next.sqrt()/alpha_t.sqrt()) + ((1/alpha_t_next - 1).sqrt() + (1/alpha_t - 1).sqrt())*noise_pred\n",
    "\n",
    "        # Store\n",
    "        intermediate_latents.append(latents)\n",
    "        if(i%250 == 0) and get_image:\n",
    "            shape = (4, H // 8, W // 8)\n",
    "            # Magic number. IDK why. Perhaps because 0.825**12<0.01 but 0.826**12>0.01\n",
    "            samples, intermediates = ddim_sampler.sample(num_inference_steps, num_samples,\n",
    "                                                        shape, cond, verbose=False, eta=eta,\n",
    "                                                        x_T=latents,\n",
    "                                                        timesteps=int(i),\n",
    "                                                        unconditional_guidance_scale=guide_scale,\n",
    "                                                        unconditional_conditioning=un_cond)\n",
    "\n",
    "            x_samples = model.decode_first_stage(samples)\n",
    "            x_samples = (einops.rearrange(x_samples, 'b c h w -> b h w c') * 127.5 + 127.5).cpu().numpy().clip(0, 255).astype(np.uint8)\n",
    "            out.append(x_samples)\n",
    "    return torch.cat(intermediate_latents),out\n",
    "def naive_invert(input_image,  model, image_resolution,\n",
    "           num_samples, a_prompt, n_prompt, prompt, ddim_sampler, control = None,guide_scale=7.5, num_inference_steps=100,\n",
    "           device='cuda',eta=1,get_image = False):\n",
    "  \n",
    "    # Encode prompt\n",
    "    input_image = HWC3(input_image)\n",
    "    img1 = resize_image(input_image, image_resolution)\n",
    "    H,W,C = img1.shape\n",
    "    ddim_sampler.make_schedule(num_inference_steps,verbose = False)\n",
    "    \n",
    "    intermediate_latents = []\n",
    "    timesteps = ddim_sampler_depth.ddim_timesteps\n",
    "    cond = { \"c_crossattn\": [model.get_learned_conditioning([prompt + ', ' + a_prompt] * num_samples)],\"c_concat\": [control] if control is not None else None}\n",
    "    un_cond = {\"c_crossattn\": [model.get_learned_conditioning([n_prompt] * num_samples)],\"c_concat\": None }\n",
    "    model.control_scales = ([1] * 13)\n",
    "\n",
    "    img1 = torch.tensor(np.array(input_image)).permute(2,0,1).unsqueeze(0).cuda()\n",
    "    L1 = model.get_first_stage_encoding(model.encode_first_stage(img1.float() / 127.5 - 1.0))\n",
    "    latents = L1\n",
    "    # Reversed timesteps <<<<<<<<<<<<<<<<<<<<\n",
    "    out = []\n",
    "    for i in range(1, num_inference_steps):\n",
    "\n",
    "        # We'll skip the final iteration\n",
    "        if i >= num_inference_steps - 1: continue\n",
    "\n",
    "        t = timesteps[i]\n",
    "        t = torch.tensor([t], dtype=torch.long, device='cuda')\n",
    "        print(t/1000, i/num_inference_steps)\n",
    "        noise = torch.randn_like(latents)\n",
    "        noisy_latents = (extract_into_tensor(model.sqrt_alphas_cumprod, t, latents.shape) * latents +\n",
    "            extract_into_tensor(model.sqrt_one_minus_alphas_cumprod, t, latents.shape) * noise)\n",
    "        noisy_latents = noisy_latents.squeeze(0)\n",
    "        noisy_latents = torch.stack([noisy_latents for _ in range(num_samples)], dim=0)\n",
    "\n",
    "        intermediate_latents.append(noisy_latents)\n",
    "        if(i%10 == 0) and get_image:\n",
    "            shape = (4, H // 8, W // 8)\n",
    "            # Magic number. IDK why. Perhaps because 0.825**12<0.01 but 0.826**12>0.01\n",
    "            samples, intermediates = ddim_sampler.sample(num_inference_steps, num_samples,\n",
    "                                                        shape, cond, verbose=False, eta=eta,\n",
    "                                                        x_T=noisy_latents,\n",
    "                                                        timesteps=int(i),\n",
    "                                                        unconditional_guidance_scale=guide_scale,\n",
    "                                                        unconditional_conditioning=un_cond)\n",
    "\n",
    "            x_samples = model.decode_first_stage(samples)\n",
    "            x_samples = (einops.rearrange(x_samples, 'b c h w -> b h w c') * 127.5 + 127.5).cpu().numpy().clip(0, 255).astype(np.uint8)\n",
    "            out.append(x_samples)\n",
    "    return torch.cat(intermediate_latents),out"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from PIL import Image\n",
    "imgg = Image.fromarray(_[0][0])\n",
    "imgg"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "source1 = cv2.imread(\"./4.png\")\n",
    "source1 = cv2.cvtColor(source1, cv2.COLOR_BGR2RGB)\n",
    "prompt = 'best qualit'\n",
    "a_prompt = '4k'\n",
    "n_prompt = 'longbody, lowres, bad anatomy, bad hands, missing fingers, pubic hair,extra digit, fewer digits, cropped, worst quality, low quality'\n",
    "out,_ = invert(input_image = source1,  model= model_depth, image_resolution= 512,\n",
    "           num_samples = 1, a_prompt=a_prompt,\n",
    "               n_prompt = n_prompt, prompt = prompt, ddim_sampler = ddim_sampler_depth, control = None,guide_scale=7.5, num_inference_steps=500,\n",
    "           device='cuda',eta = 1,get_image=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from torchvision.transforms import Resize \n",
    "def process_interpolation_add_noise_diffuse_for_loop(det, input_image1, input_image2, alpha,alpha_fram, time_frac,\n",
    "                          ddim_sampler,model, prompt, a_prompt, n_prompt, num_samples, image_resolution, detect_resolution, ddim_steps, guess_mode, strength, scale, seed, eta, low_threshold, high_threshold,clip_model,lamda, in_latend_space,grayscale):\n",
    "\n",
    "    input_image1 = HWC3(input_image1)\n",
    "    input_image2 = HWC3(input_image2)\n",
    "\n",
    "    if det == 'Canny':\n",
    "        preprocessor = CannyDetector()\n",
    "        detected_map1 = preprocessor(resize_image(input_image1, detect_resolution), low_threshold, high_threshold)\n",
    "    if det == 'Depth_Zoe':\n",
    "        preprocessor = ZoeDetector()\n",
    "        detected_map1 = preprocessor(resize_image(input_image1, detect_resolution))\n",
    "        detected_map2 = preprocessor(resize_image(input_image2, detect_resolution))\n",
    "    detected_map1 = HWC3(detected_map1)\n",
    "    detected_map2 = HWC3(detected_map2)\n",
    "    \n",
    "    img1 = resize_image(input_image1, image_resolution)\n",
    "    img2 = resize_image(input_image2, image_resolution)\n",
    "    H, W, C = img1.shape\n",
    "\n",
    "    detected_map1 = cv2.resize(detected_map1, (W, H), interpolation=cv2.INTER_LINEAR)\n",
    "    detected_map2 = cv2.resize(detected_map2, (W, H), interpolation=cv2.INTER_LINEAR)\n",
    "\n",
    "    control1 = torch.from_numpy(detected_map1.copy()).float().cuda() / 255.0\n",
    "    control2 = torch.from_numpy(detected_map2.copy()).float().cuda() / 255.0\n",
    "    control = slerp(control1,control2,alpha_fram)\n",
    "    control = torch.stack([control for _ in range(num_samples)], dim=0)\n",
    "    control = einops.rearrange(control, 'b h w c -> b c h w').clone()\n",
    "\n",
    "    if seed == -1:\n",
    "        seed = random.randint(0, 65535)\n",
    "    seed_everything(seed)\n",
    "    torch_resize = Resize([224,224])\n",
    "\n",
    "    img1 = torch.tensor(np.array(input_image1)).permute(2,0,1).unsqueeze(0).cuda()\n",
    "    img1 = img1.float() / 127.5 - 1.0\n",
    "    \n",
    "    img1_gray = img1*127.5 + 127.5\n",
    "    gray_tensor_2 = torch.sum(img1_gray * torch.tensor([0.299, 0.587, 0.114]).view(1, 3, 1, 1).cuda(), dim=1, keepdim=True)\n",
    "    \n",
    "    img2 = torch.tensor(np.array(input_image2)).permute(2,0,1).unsqueeze(0).cuda()\n",
    "    gabi  = torch_resize(img1)\n",
    "    if clip_model is not None:\n",
    "        clip_encode_img1 = clip_model.encode_image(gabi)\n",
    "    L1 = model.get_first_stage_encoding(model.encode_first_stage(img1))\n",
    "    Original_L1 = L1.detach().clone()\n",
    "    L2 = model.get_first_stage_encoding(model.encode_first_stage(img2.float() / 127.5 - 1.0))\n",
    "    T = int(time_frac* model.num_timesteps)\n",
    "    t = torch.tensor([T], dtype=torch.long, device='cuda')\n",
    "    noise = torch.randn_like(L1)\n",
    "    L1 = (extract_into_tensor(model.sqrt_alphas_cumprod, t, L1.shape) * L1 +\n",
    "        extract_into_tensor(model.sqrt_one_minus_alphas_cumprod, t, L1.shape) * noise)\n",
    "    L2 = (extract_into_tensor(model.sqrt_alphas_cumprod, t, L2.shape) * L2 +\n",
    "        extract_into_tensor(model.sqrt_one_minus_alphas_cumprod, t, L2.shape) * noise)\n",
    "    noisy_latents = slerp(L1,L2,alpha)\n",
    "    noisy_latents = noisy_latents.squeeze(0)\n",
    "    noisy_latents = torch.stack([noisy_latents for _ in range(num_samples)], dim=0)\n",
    "    noisy_latents.requires_grad_()\n",
    "    print(noisy_latents.shape)\n",
    "\n",
    "    cond = {\"c_concat\": [control], \"c_crossattn\": [model.get_learned_conditioning([prompt + ', ' + a_prompt] * num_samples)]}\n",
    "    un_cond = {\"c_concat\": None if guess_mode else [control], \"c_crossattn\": [model.get_learned_conditioning([n_prompt] * num_samples)]}\n",
    "    shape = (4, H // 8, W // 8)\n",
    "\n",
    "    model.control_scales = [strength * (0.825 ** float(12 - i)) for i in range(13)] if guess_mode else ([strength] * 13)\n",
    "    # Magic number. IDK why. Perhaps because 0.825**12<0.01 but 0.826**12>0.01\n",
    "    \n",
    "    ddim_sampler.make_schedule(ddim_steps,ddim_eta=eta,verbose = False)\n",
    "\n",
    "    intermediate_latents = []\n",
    "    timesteps = ddim_sampler_depth.ddim_timesteps\n",
    "    timesteps = timesteps[0:int(timesteps.shape[0]*time_frac)]\n",
    "    timesteps = np.flip(timesteps)\n",
    "    total_steps = timesteps.shape[0]\n",
    "    for i, step in enumerate(timesteps):\n",
    "        print(i)\n",
    "        torch.cuda.empty_cache()\n",
    "        index = total_steps - i - 1\n",
    "        t = torch.tensor([step], dtype=torch.long, device='cuda')\n",
    "        noisy_latent, pre_x0 = ddim_sampler.p_sample_ddim(noisy_latents, cond, t, index=index, unconditional_guidance_scale=scale,\n",
    "        unconditional_conditioning=un_cond)\n",
    "        if grayscale:\n",
    "            pre_x0_out = model.decode_first_stage(pre_x0)\n",
    "            pre_x0_out = pre_x0_out* 127.5 + 127.5\n",
    "            gray_tensor = torch.sum(pre_x0_out * torch.tensor([0.299, 0.587, 0.114]).view(1, 3, 1, 1).cuda(), dim=1, keepdim=True)\n",
    "\n",
    "            differece = gray_tensor - gray_tensor_2\n",
    "            norm = torch.linalg.norm(differece)\n",
    "            norm = norm.mean()\n",
    "            norm_grad = torch.autograd.grad(outputs= norm, inputs= noisy_latents)[0]\n",
    "            del pre_x0_out\n",
    "            del pre_x0\n",
    "            noisy_latents = (noisy_latent - norm_grad*lamda).detach().clone()\n",
    "            noisy_latents.requires_grad_()\n",
    "        if clip_model:\n",
    "            pre_x0_out = model.decode_first_stage(pre_x0)\n",
    "            resize_out = torch_resize(pre_x0_out)\n",
    "            clip_encoder_pre_x0 = clip_model.encode_image(resize_out)\n",
    "            differece=  clip_encode_img1 - clip_encoder_pre_x0\n",
    "            norm = torch.linalg.norm(differece)\n",
    "            norm = norm.mean()\n",
    "\n",
    "            norm_grad = torch.autograd.grad(outputs= norm, inputs= noisy_latents)[0]\n",
    "            del clip_encoder_pre_x0\n",
    "            del pre_x0_out\n",
    "            del pre_x0\n",
    "            noisy_latents = (noisy_latent - norm_grad*lamda).detach().clone()\n",
    "            noisy_latents.requires_grad_()\n",
    "        if in_latend_space:\n",
    "            differece=  pre_x0 - Original_L1\n",
    "            norm = torch.linalg.norm(differece)\n",
    "            norm = norm.mean()\n",
    "            norm_grad = torch.autograd.grad(outputs= norm, inputs= noisy_latents)[0]\n",
    "            del pre_x0\n",
    "            noisy_latents = (noisy_latent - norm_grad*lamda).detach().clone()\n",
    "            noisy_latents.requires_grad_()\n",
    "    x_samples = model.decode_first_stage(noisy_latents)\n",
    "    x_samples = (einops.rearrange(x_samples, 'b c h w -> b h w c') * 127.5 + 127.5).detach().cpu().numpy().clip(0, 255).astype(np.uint8)\n",
    "\n",
    "    results = [x_samples[i] for i in range(num_samples)]\n",
    "    return results\n",
    "def process_interpolation(det, input_image1, input_image2, alpha, time_frac,\n",
    "                          ddim_sampler,model, prompt, a_prompt, n_prompt, num_samples, image_resolution, detect_resolution, ddim_steps, guess_mode, strength, scale, seed, eta, low_threshold, high_threshold):\n",
    "    global preprocessor\n",
    "\n",
    "    with torch.no_grad():\n",
    "        input_image1 = HWC3(input_image1)\n",
    "        input_image2 = HWC3(input_image2)\n",
    "\n",
    "        if det == 'Canny':\n",
    "            preprocessor = CannyDetector()\n",
    "            detected_map1 = preprocessor(resize_image(input_image1, detect_resolution), low_threshold, high_threshold)\n",
    "        if det == 'Depth_Zoe':\n",
    "            preprocessor = ZoeDetector()\n",
    "            detected_map1 = preprocessor(resize_image(input_image1, detect_resolution))\n",
    "            detected_map2 = preprocessor(resize_image(input_image2, detect_resolution))\n",
    "        detected_map1 = HWC3(detected_map1)\n",
    "        detected_map2 = HWC3(detected_map2)\n",
    "        \n",
    "        img1 = resize_image(input_image1, image_resolution)\n",
    "        img2 = resize_image(input_image2, image_resolution)\n",
    "        H, W, C = img1.shape\n",
    "\n",
    "        detected_map1 = cv2.resize(detected_map1, (W, H), interpolation=cv2.INTER_LINEAR)\n",
    "        detected_map2 = cv2.resize(detected_map2, (W, H), interpolation=cv2.INTER_LINEAR)\n",
    "\n",
    "        control1 = torch.from_numpy(detected_map1.copy()).float().cuda() / 255.0\n",
    "        control2 = torch.from_numpy(detected_map2.copy()).float().cuda() / 255.0\n",
    "        control = slerp(control1,control2,alpha)\n",
    "        print(control.shape)\n",
    "        control = torch.stack([control for _ in range(num_samples)], dim=0)\n",
    "        control = einops.rearrange(control, 'b h w c -> b c h w').clone()\n",
    "\n",
    "        if seed == -1:\n",
    "            seed = random.randint(0, 65535)\n",
    "        seed_everything(seed)\n",
    "\n",
    "\n",
    "        img1 = torch.tensor(np.array(input_image1)).permute(2,0,1).unsqueeze(0).cuda()\n",
    "        img2 = torch.tensor(np.array(input_image2)).permute(2,0,1).unsqueeze(0).cuda()\n",
    "        L1 = model.get_first_stage_encoding(model.encode_first_stage(img1.float() / 127.5 - 1.0))\n",
    "        L2 = model.get_first_stage_encoding(model.encode_first_stage(img2.float() / 127.5 - 1.0))\n",
    "        latents = slerp(L1,L2,alpha)\n",
    "        T = int(time_frac* model.num_timesteps)\n",
    "        t = torch.tensor([T], dtype=torch.long, device='cuda')\n",
    "        noise = torch.randn_like(latents)\n",
    "        noisy_latents = (extract_into_tensor(model.sqrt_alphas_cumprod, t, latents.shape) * latents +\n",
    "            extract_into_tensor(model.sqrt_one_minus_alphas_cumprod, t, latents.shape) * noise)\n",
    "        noisy_latents = noisy_latents.squeeze(0)\n",
    "        noisy_latents = torch.stack([noisy_latents for _ in range(num_samples)], dim=0)\n",
    "        print(noisy_latents.shape)\n",
    "\n",
    "        cond = {\"c_concat\": [control], \"c_crossattn\": [model.get_learned_conditioning([prompt + ', ' + a_prompt] * num_samples)]}\n",
    "        un_cond = {\"c_concat\": None if guess_mode else [control], \"c_crossattn\": [model.get_learned_conditioning([n_prompt] * num_samples)]}\n",
    "        shape = (4, H // 8, W // 8)\n",
    "\n",
    "        model.control_scales = [strength * (0.825 ** float(12 - i)) for i in range(13)] if guess_mode else ([strength] * 13)\n",
    "        # Magic number. IDK why. Perhaps because 0.825**12<0.01 but 0.826**12>0.01\n",
    "\n",
    "        samples, intermediates = ddim_sampler.sample(ddim_steps, num_samples,\n",
    "                                                     shape, cond, verbose=False, eta=eta,\n",
    "                                                     x_T=noisy_latents,\n",
    "                                                     timesteps=int(time_frac * ddim_steps),\n",
    "                                                     unconditional_guidance_scale=scale,\n",
    "                                                     unconditional_conditioning=un_cond)\n",
    "\n",
    "        x_samples = model.decode_first_stage(samples)\n",
    "        x_samples = (einops.rearrange(x_samples, 'b c h w -> b h w c') * 127.5 + 127.5).cpu().numpy().clip(0, 255).astype(np.uint8)\n",
    "\n",
    "        results = [x_samples[i] for i in range(num_samples)]\n",
    "    return results\n",
    "def process_interpolation_add_noise_diffuse(det, input_image1, input_image2, alpha,alpha_fram, time_frac,\n",
    "                          ddim_sampler,model, prompt, a_prompt, n_prompt, num_samples, image_resolution, detect_resolution, ddim_steps, guess_mode, strength, scale, seed, eta, low_threshold, high_threshold,only_language = False):\n",
    "    global preprocessor\n",
    "\n",
    "    with torch.no_grad():\n",
    "        input_image1 = HWC3(input_image1)\n",
    "        input_image2 = HWC3(input_image2)\n",
    "\n",
    "        if det == 'Canny':\n",
    "            preprocessor = CannyDetector()\n",
    "            detected_map1 = preprocessor(resize_image(input_image1, detect_resolution), low_threshold, high_threshold)\n",
    "            detected_map2 = preprocessor(resize_image(input_image2, detect_resolution), low_threshold, high_threshold)\n",
    "\n",
    "        if det == 'Depth_Zoe':\n",
    "            preprocessor = ZoeDetector()\n",
    "            detected_map1 = preprocessor(resize_image(input_image1, detect_resolution))\n",
    "            detected_map2 = preprocessor(resize_image(input_image2, detect_resolution))\n",
    "        detected_map1 = HWC3(detected_map1)\n",
    "        detected_map2 = HWC3(detected_map2)\n",
    "        \n",
    "        img1 = resize_image(input_image1, image_resolution)\n",
    "        img2 = resize_image(input_image2, image_resolution)\n",
    "        H, W, C = img1.shape\n",
    "\n",
    "        detected_map1 = cv2.resize(detected_map1, (W, H), interpolation=cv2.INTER_LINEAR)\n",
    "        detected_map2 = cv2.resize(detected_map2, (W, H), interpolation=cv2.INTER_LINEAR)\n",
    "\n",
    "        control1 = torch.from_numpy(detected_map1.copy()).float().cuda() / 255.0\n",
    "        control2 = torch.from_numpy(detected_map2.copy()).float().cuda() / 255.0\n",
    "        control = slerp(control1,control2,alpha_fram)\n",
    "        print(control.shape)\n",
    "        control = torch.stack([control for _ in range(num_samples)], dim=0)\n",
    "        control = einops.rearrange(control, 'b h w c -> b c h w').clone()\n",
    "\n",
    "        if seed == -1:\n",
    "            seed = random.randint(0, 65535)\n",
    "        seed_everything(seed)\n",
    "\n",
    "\n",
    "        img1 = torch.tensor(np.array(input_image1)).permute(2,0,1).unsqueeze(0).cuda()\n",
    "        img2 = torch.tensor(np.array(input_image2)).permute(2,0,1).unsqueeze(0).cuda()\n",
    "        L1 = model.get_first_stage_encoding(model.encode_first_stage(img1.float() / 127.5 - 1.0))\n",
    "        L2 = model.get_first_stage_encoding(model.encode_first_stage(img2.float() / 127.5 - 1.0))\n",
    "        T = int(time_frac* model.num_timesteps)\n",
    "        t = torch.tensor([T], dtype=torch.long, device='cuda')\n",
    "        noise = torch.randn_like(L1)\n",
    "        L1 = (extract_into_tensor(model.sqrt_alphas_cumprod, t, L1.shape) * L1 +\n",
    "            extract_into_tensor(model.sqrt_one_minus_alphas_cumprod, t, L1.shape) * noise)\n",
    "        L2 = (extract_into_tensor(model.sqrt_alphas_cumprod, t, L2.shape) * L2 +\n",
    "            extract_into_tensor(model.sqrt_one_minus_alphas_cumprod, t, L2.shape) * noise)\n",
    "        noisy_latents = slerp(L1,L2,alpha)\n",
    "        noisy_latents = noisy_latents.squeeze(0)\n",
    "        noisy_latents = torch.stack([noisy_latents for _ in range(num_samples)], dim=0)\n",
    "        print(noisy_latents.shape)\n",
    "\n",
    "        cond = {\"c_concat\": [control], \"c_crossattn\": [model.get_learned_conditioning([prompt + ', ' + a_prompt] * num_samples)]}\n",
    "        un_cond = {\"c_concat\": None if guess_mode else [control], \"c_crossattn\": [model.get_learned_conditioning([n_prompt] * num_samples)]}\n",
    "        if only_language:\n",
    "            cond = {\"c_crossattn\": [model.get_learned_conditioning([prompt + ', ' + a_prompt] * num_samples)],\"c_concat\": None }\n",
    "            un_cond = {\"c_crossattn\": [model.get_learned_conditioning([n_prompt] * num_samples)],\"c_concat\": None}            \n",
    "        shape = (4, H // 8, W // 8)\n",
    "\n",
    "        model.control_scales = [strength * (0.825 ** float(12 - i)) for i in range(13)] if guess_mode else ([strength] * 13)\n",
    "        # Magic number. IDK why. Perhaps because 0.825**12<0.01 but 0.826**12>0.01\n",
    "\n",
    "        samples, intermediates = ddim_sampler.sample(ddim_steps, num_samples,\n",
    "                                                     shape, cond, verbose=False, eta=eta,\n",
    "                                                     x_T=noisy_latents,\n",
    "                                                     timesteps=int(time_frac * ddim_steps),\n",
    "                                                     unconditional_guidance_scale=scale,\n",
    "                                                     unconditional_conditioning=un_cond)\n",
    "\n",
    "        x_samples = model.decode_first_stage(samples)\n",
    "        x_samples = (einops.rearrange(x_samples, 'b c h w -> b h w c') * 127.5 + 127.5).cpu().numpy().clip(0, 255).astype(np.uint8)\n",
    "\n",
    "        results = [x_samples[i] for i in range(num_samples)]\n",
    "    return results\n",
    "def process_interpolation_add_noise_diffuse_with_language_interpolation(det, input_image1, input_image2, alpha, time_frac, word1,word2,\n",
    "                          ddim_sampler,model, prompt, a_prompt, n_prompt, num_samples, image_resolution, detect_resolution, ddim_steps, guess_mode, strength, scale, seed, eta, low_threshold, high_threshold):\n",
    "    global preprocessor\n",
    "\n",
    "    with torch.no_grad():\n",
    "        input_image1 = HWC3(input_image1)\n",
    "        input_image2 = HWC3(input_image2)\n",
    "\n",
    "        if det == 'Canny':\n",
    "            preprocessor = CannyDetector()\n",
    "            detected_map1 = preprocessor(resize_image(input_image1, detect_resolution), low_threshold, high_threshold)\n",
    "        if det == 'Depth_Zoe':\n",
    "            preprocessor = ZoeDetector()\n",
    "            detected_map1 = preprocessor(resize_image(input_image1, detect_resolution))\n",
    "            detected_map2 = preprocessor(resize_image(input_image2, detect_resolution))\n",
    "        detected_map1 = HWC3(detected_map1)\n",
    "        detected_map2 = HWC3(detected_map2)\n",
    "        \n",
    "        img1 = resize_image(input_image1, image_resolution)\n",
    "        img2 = resize_image(input_image2, image_resolution)\n",
    "        H, W, C = img1.shape\n",
    "\n",
    "        detected_map1 = cv2.resize(detected_map1, (W, H), interpolation=cv2.INTER_LINEAR)\n",
    "        detected_map2 = cv2.resize(detected_map2, (W, H), interpolation=cv2.INTER_LINEAR)\n",
    "\n",
    "        control1 = torch.from_numpy(detected_map1.copy()).float().cuda() / 255.0\n",
    "        control2 = torch.from_numpy(detected_map2.copy()).float().cuda() / 255.0\n",
    "        control = slerp(control1,control2,alpha)\n",
    "        print(control.shape)\n",
    "        control = torch.stack([control for _ in range(num_samples)], dim=0)\n",
    "        control = einops.rearrange(control, 'b h w c -> b c h w').clone()\n",
    "\n",
    "        if seed == -1:\n",
    "            seed = random.randint(0, 65535)\n",
    "        seed_everything(seed)\n",
    "\n",
    "\n",
    "        img1 = torch.tensor(np.array(input_image1)).permute(2,0,1).unsqueeze(0).cuda()\n",
    "        img2 = torch.tensor(np.array(input_image2)).permute(2,0,1).unsqueeze(0).cuda()\n",
    "        L1 = model.get_first_stage_encoding(model.encode_first_stage(img1.float() / 127.5 - 1.0))\n",
    "        L2 = model.get_first_stage_encoding(model.encode_first_stage(img2.float() / 127.5 - 1.0))\n",
    "        T = int(time_frac* model.num_timesteps)\n",
    "        t = torch.tensor([T], dtype=torch.long, device='cuda')\n",
    "        noise = torch.randn_like(L1)\n",
    "        L1 = (extract_into_tensor(model.sqrt_alphas_cumprod, t, L1.shape) * L1 +\n",
    "            extract_into_tensor(model.sqrt_one_minus_alphas_cumprod, t, L1.shape) * noise)\n",
    "        L2 = (extract_into_tensor(model.sqrt_alphas_cumprod, t, L2.shape) * L2 +\n",
    "            extract_into_tensor(model.sqrt_one_minus_alphas_cumprod, t, L2.shape) * noise)\n",
    "        noisy_latents = slerp(L1,L2,alpha)\n",
    "        noisy_latents = noisy_latents.squeeze(0)\n",
    "        noisy_latents = torch.stack([noisy_latents for _ in range(num_samples)], dim=0)\n",
    "        print(noisy_latents.shape)\n",
    "        tep1 = model.get_learned_conditioning([prompt + ', ' + a_prompt + ',' + word1] * num_samples)\n",
    "        tep2 = model.get_learned_conditioning([prompt + ', ' + a_prompt + ',' + word2] * num_samples)\n",
    "        cond = {\"c_concat\": [control], \"c_crossattn\": [slerp(tep1,tep2,alpha)]}\n",
    "        un_cond = {\"c_concat\": None if guess_mode else [control], \"c_crossattn\": [model.get_learned_conditioning([n_prompt] * num_samples)]}\n",
    "        shape = (4, H // 8, W // 8)\n",
    "\n",
    "        model.control_scales = [strength * (0.825 ** float(12 - i)) for i in range(13)] if guess_mode else ([strength] * 13)\n",
    "        # Magic number. IDK why. Perhaps because 0.825**12<0.01 but 0.826**12>0.01\n",
    "\n",
    "        samples, intermediates = ddim_sampler.sample(ddim_steps, num_samples,\n",
    "                                                     shape, cond, verbose=False, eta=eta,\n",
    "                                                     x_T=noisy_latents,\n",
    "                                                     timesteps=int(time_frac * ddim_steps),\n",
    "                                                     unconditional_guidance_scale=scale,\n",
    "                                                     unconditional_conditioning=un_cond)\n",
    "\n",
    "        x_samples = model.decode_first_stage(samples)\n",
    "        x_samples = (einops.rearrange(x_samples, 'b c h w -> b h w c') * 127.5 + 127.5).cpu().numpy().clip(0, 255).astype(np.uint8)\n",
    "\n",
    "        results = [x_samples[i] for i in range(num_samples)]\n",
    "    return results\n",
    "def process_interpolation_add_noise_diffuse_with_language_interpolation_with_middle_lnguage(det, input_image1, input_image2,detect_image, alpha, time_frac, word1,word2,\n",
    "                          ddim_sampler,model, prompt, a_prompt, n_prompt, num_samples, image_resolution, detect_resolution, ddim_steps, guess_mode, strength, scale, seed, eta, low_threshold, high_threshold):\n",
    "    global preprocessor\n",
    "\n",
    "    with torch.no_grad():\n",
    "        input_image1 = HWC3(input_image1)\n",
    "        input_image2 = HWC3(input_image2)\n",
    "        img1 = resize_image(input_image1, image_resolution)\n",
    "        img2 = resize_image(input_image2, image_resolution)\n",
    "        H, W, C = img1.shape\n",
    "        if detect_image is None:\n",
    "            if det == 'Canny':\n",
    "                preprocessor = CannyDetector()\n",
    "                detected_map1 = preprocessor(resize_image(input_image1, detect_resolution), low_threshold, high_threshold)\n",
    "            if det == 'Depth_Zoe':\n",
    "                preprocessor = ZoeDetector()\n",
    "                detected_map1 = preprocessor(resize_image(input_image1, detect_resolution))\n",
    "                detected_map2 = preprocessor(resize_image(input_image2, detect_resolution))\n",
    "            \n",
    "            detected_map1 = HWC3(detected_map1)\n",
    "            detected_map2 = HWC3(detected_map2)\n",
    "\n",
    "            detected_map1 = cv2.resize(detected_map1, (W, H), interpolation=cv2.INTER_LINEAR)\n",
    "            detected_map2 = cv2.resize(detected_map2, (W, H), interpolation=cv2.INTER_LINEAR)\n",
    "\n",
    "            control1 = torch.from_numpy(detected_map1.copy()).float().cuda() / 255.0\n",
    "            control2 = torch.from_numpy(detected_map2.copy()).float().cuda() / 255.0\n",
    "            control = slerp(control1,control2,alpha)\n",
    "            print(control.shape)\n",
    "            control = torch.stack([control for _ in range(num_samples)], dim=0)\n",
    "            control = einops.rearrange(control, 'b h w c -> b c h w').clone()\n",
    "        else:\n",
    "            detect_image = HWC3(detect_image)\n",
    "            preprocessor = ZoeDetector()\n",
    "            detected_map1 = preprocessor(resize_image(detect_image, detect_resolution))\n",
    "            detected_map1 = HWC3(detected_map1)\n",
    "            detected_map1 = cv2.resize(detected_map1, (W, H), interpolation=cv2.INTER_LINEAR)\n",
    "            control = torch.from_numpy(detected_map1.copy()).float().cuda() / 255.0\n",
    "            control = torch.stack([control for _ in range(num_samples)], dim=0)\n",
    "            control = einops.rearrange(control, 'b h w c -> b c h w').clone()\n",
    "        if seed == -1:\n",
    "            seed = random.randint(0, 65535)\n",
    "        seed_everything(seed)\n",
    "\n",
    "\n",
    "        img1 = torch.tensor(np.array(input_image1)).permute(2,0,1).unsqueeze(0).cuda()\n",
    "        img2 = torch.tensor(np.array(input_image2)).permute(2,0,1).unsqueeze(0).cuda()\n",
    "        L1 = model.get_first_stage_encoding(model.encode_first_stage(img1.float() / 127.5 - 1.0))\n",
    "        L2 = model.get_first_stage_encoding(model.encode_first_stage(img2.float() / 127.5 - 1.0))\n",
    "        T = int(time_frac* model.num_timesteps)\n",
    "        t = torch.tensor([T], dtype=torch.long, device='cuda')\n",
    "        noise = torch.randn_like(L1)\n",
    "        L1 = (extract_into_tensor(model.sqrt_alphas_cumprod, t, L1.shape) * L1 +\n",
    "            extract_into_tensor(model.sqrt_one_minus_alphas_cumprod, t, L1.shape) * noise)\n",
    "        L2 = (extract_into_tensor(model.sqrt_alphas_cumprod, t, L2.shape) * L2 +\n",
    "            extract_into_tensor(model.sqrt_one_minus_alphas_cumprod, t, L2.shape) * noise)\n",
    "        noisy_latents = slerp(L1,L2,alpha)\n",
    "        noisy_latents = noisy_latents.squeeze(0)\n",
    "        noisy_latents = torch.stack([noisy_latents for _ in range(num_samples)], dim=0)\n",
    "        print(noisy_latents.shape)\n",
    "        tep1 = model.get_learned_conditioning([prompt + ', ' + a_prompt + ',' + word1 + ' and ' + word2] * num_samples)\n",
    "        tep2 = model.get_learned_conditioning([prompt + ', ' + a_prompt + ',' + word1] * num_samples)\n",
    "        tep3 = model.get_learned_conditioning([prompt + ', ' + a_prompt + ',' + word2] * num_samples)\n",
    "        if alpha <= 0.5:   \n",
    "            cond = {\"c_concat\": [control], \"c_crossattn\": [slerp(tep2,tep1,alpha/0.5)]}\n",
    "        else:\n",
    "            cond = {\"c_concat\": [control], \"c_crossattn\": [slerp(tep1,tep3,(alpha - 0.5)/0.5)]}\n",
    "\n",
    "        un_cond = {\"c_concat\": None if guess_mode else [control], \"c_crossattn\": [model.get_learned_conditioning([n_prompt] * num_samples)]}\n",
    "        shape = (4, H // 8, W // 8)\n",
    "\n",
    "        model.control_scales = [strength * (0.825 ** float(12 - i)) for i in range(13)] if guess_mode else ([strength] * 13)\n",
    "        # Magic number. IDK why. Perhaps because 0.825**12<0.01 but 0.826**12>0.01\n",
    "\n",
    "        samples, intermediates = ddim_sampler.sample(ddim_steps, num_samples,\n",
    "                                                     shape, cond, verbose=False, eta=eta,\n",
    "                                                     x_T=noisy_latents,\n",
    "                                                     timesteps=int(time_frac * ddim_steps),\n",
    "                                                     unconditional_guidance_scale=scale,\n",
    "                                                     unconditional_conditioning=un_cond)\n",
    "\n",
    "        x_samples = model.decode_first_stage(samples)\n",
    "        x_samples = (einops.rearrange(x_samples, 'b c h w -> b h w c') * 127.5 + 127.5).cpu().numpy().clip(0, 255).astype(np.uint8)\n",
    "\n",
    "        results = [x_samples[i] for i in range(num_samples)]\n",
    "    return results\n",
    "def style_transfer_fixing_structure(det, input_image1, detect_image, time_frac,noisylatend,\n",
    "                          ddim_sampler,model, prompt, a_prompt, n_prompt, num_samples, image_resolution, detect_resolution, ddim_steps, guess_mode, strength, scale, seed, eta, low_threshold, high_threshold):\n",
    "    global preprocessor\n",
    "\n",
    "    with torch.no_grad():\n",
    "        input_image1 = HWC3(input_image1)\n",
    "\n",
    "        if det == 'Canny':\n",
    "            preprocessor = CannyDetector()\n",
    "            detected_map1 = preprocessor(resize_image(input_image1, detect_resolution), low_threshold, high_threshold)\n",
    "        if det == 'Depth_Zoe':\n",
    "            preprocessor = ZoeDetector()\n",
    "            detected_map1 = preprocessor(resize_image(input_image1, detect_resolution))\n",
    "        detected_map1 = HWC3(detected_map1)\n",
    "        \n",
    "        img1 = resize_image(input_image1, image_resolution)\n",
    "        H, W, C = img1.shape\n",
    "\n",
    "        detect_image = HWC3(detect_image)\n",
    "        preprocessor = ZoeDetector()\n",
    "        detected_map1 = preprocessor(resize_image(detect_image, detect_resolution))\n",
    "        detected_map1 = HWC3(detected_map1)\n",
    "        detected_map1 = cv2.resize(detected_map1, (W, H), interpolation=cv2.INTER_LINEAR)\n",
    "        control = torch.from_numpy(detected_map1.copy()).float().cuda() / 255.0\n",
    "        control = torch.stack([control for _ in range(num_samples)], dim=0)\n",
    "        control = einops.rearrange(control, 'b h w c -> b c h w').clone()\n",
    "\n",
    "\n",
    "        if seed == -1:\n",
    "            seed = random.randint(0, 65535)\n",
    "        seed_everything(seed)\n",
    "\n",
    "\n",
    "        img1 = torch.tensor(np.array(input_image1)).permute(2,0,1).unsqueeze(0).cuda()\n",
    "        L1 = model.get_first_stage_encoding(model.encode_first_stage(img1.float() / 127.5 - 1.0))\n",
    "        T = int(time_frac* model.num_timesteps)\n",
    "        t = torch.tensor([T], dtype=torch.long, device='cuda')\n",
    "        noise = torch.randn_like(L1)\n",
    "        L1 = (extract_into_tensor(model.sqrt_alphas_cumprod, t, L1.shape) * L1 +\n",
    "            extract_into_tensor(model.sqrt_one_minus_alphas_cumprod, t, L1.shape) * noise)\n",
    "        noisy_latents = L1\n",
    "        noisy_latents = noisy_latents.squeeze(0)\n",
    "        noisy_latents = torch.stack([noisy_latents for _ in range(num_samples)], dim=0)\n",
    "        print(noisy_latents.shape)\n",
    "\n",
    "        cond = {\"c_concat\": [control], \"c_crossattn\": [model.get_learned_conditioning([prompt + ', ' + a_prompt] * num_samples)]}\n",
    "        un_cond = {\"c_concat\": None if guess_mode else [control], \"c_crossattn\": [model.get_learned_conditioning([n_prompt] * num_samples)]}\n",
    "        shape = (4, H // 8, W // 8)\n",
    "\n",
    "        model.control_scales = [strength * (0.825 ** float(12 - i)) for i in range(13)] if guess_mode else ([strength] * 13)\n",
    "        # Magic number. IDK why. Perhaps because 0.825**12<0.01 but 0.826**12>0.01\n",
    "        if noisylatend is not None :\n",
    "            noisylatend = torch.stack([noisylatend for _ in range(num_samples)], dim=0)\n",
    "            samples, intermediates = ddim_sampler.sample(ddim_steps, num_samples,\n",
    "                                                        shape, cond, verbose=False, eta=eta,\n",
    "                                                        x_T=noisylatend,\n",
    "                                                        timesteps=int(time_frac * ddim_steps),\n",
    "                                                        unconditional_guidance_scale=scale,\n",
    "                                                        unconditional_conditioning=un_cond)\n",
    "        else:\n",
    "            samples, intermediates = ddim_sampler.sample(ddim_steps, num_samples,\n",
    "                                                        shape, cond, verbose=False, eta=eta,\n",
    "                                                        x_T=noisy_latents,\n",
    "                                                        timesteps=int(time_frac * ddim_steps),\n",
    "                                                        unconditional_guidance_scale=scale,\n",
    "                                                        unconditional_conditioning=un_cond)\n",
    "        x_samples = model.decode_first_stage(samples)\n",
    "        x_samples = (einops.rearrange(x_samples, 'b c h w -> b h w c') * 127.5 + 127.5).cpu().numpy().clip(0, 255).astype(np.uint8)\n",
    "\n",
    "        results = [x_samples[i] for i in range(num_samples)]\n",
    "    return results\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "source1 = cv2.imread(\"./4.png\")\n",
    "source1 = cv2.cvtColor(source1, cv2.COLOR_BGR2RGB)\n",
    "source2 = cv2.imread(\"./5.png\")\n",
    "source2 = cv2.cvtColor(source2, cv2.COLOR_BGR2RGB)\n",
    "prompt = 'best qualit'\n",
    "a_prompt = '4k'\n",
    "n_prompt = 'longbody, lowres, bad anatomy, bad hands, missing fingers, pubic hair,extra digit, fewer digits, cropped, worst quality, low quality'\n",
    "n = 10\n",
    "for i in range(8):\n",
    "    for j in range(6):\n",
    "        os.makedirs(\"./Result/comb_16/\",exist_ok=True)\n",
    "        out = process_interpolation_add_noise_diffuse(det='Depth_Zoe',input_image1=source1,input_image2=source2, alpha= i*0.05 ,alpha_fram=1, time_frac=0.05*j + 0.6, \n",
    "                                    ddim_sampler=ddim_sampler_depth, model= model_depth, a_prompt=a_prompt,\n",
    "                                    n_prompt = n_prompt, num_samples = 1, \n",
    "                                    image_resolution = 512, prompt=prompt, detect_resolution=512, ddim_steps = 100, guess_mode =None, strength = 1, scale = 5, seed = 4, eta = 1, low_threshold = 100, high_threshold = 200,grayscale = True)\n",
    "        Image.fromarray(out[0]).save(f'./Result/comb_16/{i:03d}{j:03d}.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "torch.cuda.empty_cache()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "source1 = cv2.imread(\"./4.png\")\n",
    "source1 = cv2.cvtColor(source1, cv2.COLOR_BGR2RGB)\n",
    "source2 = cv2.imread(\"./5.png\")\n",
    "source2 = cv2.cvtColor(source2, cv2.COLOR_BGR2RGB)\n",
    "prompt = 'best qualit'\n",
    "a_prompt = '4k'\n",
    "n_prompt = 'longbody, lowres, bad anatomy, bad hands, missing fingers, pubic hair,extra digit, fewer digits, cropped, worst quality, low quality'\n",
    "n = 10\n",
    "for i in range(8):\n",
    "    for j in range(10):\n",
    "        if i < 1 and j <= 9:\n",
    "            continue\n",
    "        print(i,j)\n",
    "        os.makedirs(\"./Result/comb_18/\",exist_ok=True)\n",
    "        out = process_interpolation_add_noise_diffuse_for_loop(det='Depth_Zoe',input_image1=source1,input_image2=source2, alpha= 0,alpha_fram=1, time_frac=0.55 + 0.04*i, \n",
    "                                    ddim_sampler=ddim_sampler_depth, model= model_depth, a_prompt=a_prompt,\n",
    "                                    n_prompt = n_prompt, num_samples = 1, \n",
    "                                    image_resolution = 512, prompt=prompt, detect_resolution=512, ddim_steps = 100, guess_mode =None, strength = 1, scale = 5, seed = 4, eta = 1, low_threshold = 100, high_threshold = 200,clip_model=None,lamda = 0.1*j,in_latend_space=True )\n",
    "        Image.fromarray(out[0]).save(f'./Result/comb_18/{i:03d}{j:03d}.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "img_size [384, 512]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Params passed to Resize transform:\n",
      "\twidth:  512\n",
      "\theight:  384\n",
      "\tresize_target:  True\n",
      "\tkeep_aspect_ratio:  True\n",
      "\tensure_multiple_of:  32\n",
      "\tresize_method:  minimal\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Global seed set to 4\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([1, 4, 64, 64])\n",
      "0\n",
      "1\n",
      "2\n",
      "3\n",
      "4\n",
      "5\n",
      "6\n",
      "7\n",
      "8\n",
      "9\n",
      "10\n",
      "11\n",
      "12\n",
      "13\n",
      "14\n",
      "15\n",
      "16\n",
      "17\n",
      "18\n",
      "19\n",
      "20\n",
      "21\n",
      "22\n",
      "23\n",
      "24\n",
      "25\n",
      "26\n",
      "27\n",
      "28\n",
      "29\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "source1 = cv2.imread(\"./4.png\")\n",
    "source1 = cv2.cvtColor(source1, cv2.COLOR_BGR2RGB)\n",
    "source2 = cv2.imread(\"./5.png\")\n",
    "source2 = cv2.cvtColor(source2, cv2.COLOR_BGR2RGB)\n",
    "prompt = 'best qualit'\n",
    "a_prompt = '4k'\n",
    "n_prompt = 'longbody, lowres, bad anatomy, bad hands, missing fingers, pubic hair,extra digit, fewer digits, cropped, worst quality, low quality'\n",
    "\n",
    "out = process_interpolation_add_noise_diffuse_for_loop(det='Depth_Zoe',input_image1=source1,input_image2=source2, alpha= 1,alpha_fram=0, time_frac=0.3, \n",
    "                                    ddim_sampler=ddim_sampler_depth, model= model_depth, a_prompt=a_prompt,\n",
    "                                    n_prompt = n_prompt, num_samples = 1, \n",
    "                                    image_resolution = 512, prompt=prompt, detect_resolution=512, ddim_steps = 100, guess_mode =None, strength = 1, scale = 5, seed = 4, eta = 1, low_threshold = 100, high_threshold = 200,clip_model=None,lamda = 0.01,in_latend_space= None,grayscale=True )\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<PIL.Image.Image image mode=RGB size=512x512>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from PIL import Image\n",
    "imgg = Image.fromarray(out[0])\n",
    "imgg"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i in range(10):\n",
    "    os.makedirs(\"./Result/comb_17_time_frac0.7/\",exist_ok=True)\n",
    "    Image.fromarray(yaya[i][0]).save(f'./Result/comb_17_time_frac0.7/{i:03d}.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from PIL import Image\n",
    "imgg = Image.fromarray(out[0])\n",
    "imgg.save(\"Result/interpolation/0_05.png\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "out = resize_image_to_squra(\"./test_imgs/h001.png\",512)\n",
    "Image.fromarray(out).save(\"./4.png\")\n",
    "out = resize_image_to_squra(\"./test_imgs/h002.png\",512)\n",
    "Image.fromarray(out).save(\"./5.png\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "source1 = cv2.imread(\"./4.png\")\n",
    "source1 = cv2.cvtColor(source1, cv2.COLOR_BGR2RGB)\n",
    "source2 = cv2.imread(\"./5.png\")\n",
    "source2 = cv2.cvtColor(source2, cv2.COLOR_BGR2RGB)\n",
    "prompt = 'best qualit'\n",
    "a_prompt = '4k'\n",
    "n_prompt = 'longbody, lowres, bad anatomy, bad hands, missing fingers, pubic hair,extra digit, fewer digits, cropped, worst quality, low quality'\n",
    "\n",
    "out = process_interpolation_add_noise_diffuse(det='Depth_Zoe',input_image1=source1,input_image2=source2, alpha= 1,alpha_fram=0, time_frac=0.8, \n",
    "                                ddim_sampler=ddim_sampler_depth, model= model_depth, a_prompt=a_prompt,\n",
    "                                n_prompt = n_prompt, num_samples = 1, \n",
    "                                image_resolution = 512, prompt=prompt, detect_resolution=512, ddim_steps = 75, guess_mode =None, strength = 1, scale = 5, seed = 4, eta = 1, low_threshold = 100, high_threshold = 200,only_language= False )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "source1 = cv2.imread(\"./4.png\")\n",
    "source1 = cv2.cvtColor(source1, cv2.COLOR_BGR2RGB)\n",
    "source2 = cv2.imread(\"./0.png\")\n",
    "source2 = cv2.cvtColor(source2, cv2.COLOR_BGR2RGB)\n",
    "prompt = 'best qualit'\n",
    "a_prompt = '4k'\n",
    "n_prompt = 'longbody, lowres, bad anatomy, bad hands, missing fingers, pubic hair,extra digit, fewer digits, cropped, worst quality, low quality'\n",
    "\n",
    "out = process_interpolation_add_noise_diffuse(det='Canny',input_image1=source1,input_image2=source2, alpha= 0,alpha_fram=1, time_frac=0.7, \n",
    "                                ddim_sampler=ddim_sampler, model= model, a_prompt=a_prompt,\n",
    "                                n_prompt = n_prompt, num_samples = 1, \n",
    "                                image_resolution = 512, prompt=prompt, detect_resolution=512, ddim_steps = 100, guess_mode =None, strength = 1, scale = 5, seed = 4, eta = 1, low_threshold = 100, high_threshold = 200,only_language= False )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "source1 = cv2.imread(\"./4.png\")\n",
    "source1 = cv2.cvtColor(source1, cv2.COLOR_BGR2RGB)\n",
    "source2 = cv2.imread(\"./5.png\")\n",
    "source2 = cv2.cvtColor(source2, cv2.COLOR_BGR2RGB)\n",
    "prompt = 'best qualit, black person and poker'\n",
    "a_prompt = '4k'\n",
    "n_prompt = 'longbody, lowres, bad anatomy, bad hands, missing fingers, pubic hair,extra digit, fewer digits, cropped, worst quality, low quality'\n",
    "\n",
    "out = process_interpolation(det='Depth_Zoe',input_image1=source1,input_image2=source2, alpha= 0.5 ,time_frac=0.85,\n",
    "                                ddim_sampler=ddim_sampler_depth, model= model_depth, a_prompt=a_prompt,\n",
    "                                n_prompt = n_prompt, num_samples = 4, \n",
    "                                image_resolution = 512, prompt=prompt, detect_resolution=512, ddim_steps = 100, guess_mode =None, strength = 1, scale = 5, seed = 4, eta = 1, low_threshold = 100, high_threshold = 200 )\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "source1 = cv2.imread(\"./4.png\")\n",
    "source1 = cv2.cvtColor(source1, cv2.COLOR_BGR2RGB)\n",
    "source2 = cv2.imread(\"./5.png\")\n",
    "source2 = cv2.cvtColor(source2, cv2.COLOR_BGR2RGB)\n",
    "prompt = 'best qualit'\n",
    "a_prompt = '4k'\n",
    "n_prompt = 'longbody, lowres, bad anatomy, bad hands, missing fingers, pubic hair,extra digit, fewer digits, cropped, worst quality, low quality'\n",
    "word1 = 'clown'\n",
    "word2 = 'Soldier'\n",
    "for i in range(31):\n",
    "    os.makedirs(\"./Result/comb_10_with_middle_lnguage_030/\",exist_ok=True)\n",
    "    out = process_interpolation_add_noise_diffuse_with_language_interpolation_with_middle_lnguage(det='Depth_Zoe',input_image1=source1,input_image2=source2, alpha= i/30 ,\n",
    "                                word1=word1,word2=word2,time_frac=0.30,detect_image = None,\n",
    "                                ddim_sampler=ddim_sampler_depth, model= model_depth, a_prompt=a_prompt,\n",
    "                                n_prompt = n_prompt, num_samples = 1, \n",
    "                                image_resolution = 512, prompt=prompt, detect_resolution=512, ddim_steps = 100, guess_mode =None, strength = 1, scale = 5, seed = 4, eta = 1, low_threshold = 100, high_threshold = 200 )\n",
    "    Image.fromarray(out[0]).save(f'./Result/comb_10_with_middle_lnguage_030/{i:03d}.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "out.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "from PIL import Image\n",
    "\n",
    "source1 = cv2.imread(\"./5.png\")\n",
    "source1 = cv2.cvtColor(source1, cv2.COLOR_BGR2RGB)\n",
    "source2 = cv2.imread(\"./4.png\")\n",
    "source2 = cv2.cvtColor(source2, cv2.COLOR_BGR2RGB)\n",
    "prompt = 'best qualit.the great wall'\n",
    "a_prompt = '4k'\n",
    "n_prompt = 'longbody, lowres, bad anatomy, bad hands, missing fingers, pubic hair,extra digit, fewer digits, cropped, worst quality, low quality'\n",
    "os.makedirs(\"./Result/style5\",exist_ok=True)\n",
    "\n",
    "for i in range(10):\n",
    "    ou = style_transfer_fixing_structure(det='Depth_Zoe',input_image1=source2,\n",
    "                                time_frac=0.05*i + 0.5,detect_image = source1,noisylatend=None,\n",
    "                                ddim_sampler=ddim_sampler_depth, model= model_depth, a_prompt=a_prompt,\n",
    "                                n_prompt = n_prompt, num_samples = 1, \n",
    "                                image_resolution = 512, prompt=prompt, detect_resolution=512, ddim_steps = 100, guess_mode =None, strength = 1, scale = 5, seed = 4, eta = 1, low_threshold = 100, high_threshold = 200 )\n",
    "    Image.fromarray(ou[0]).save(f'./Result/style5/with_promt{i:03d}.png')\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "source1 = cv2.imread(\"./4.png\")\n",
    "source1 = cv2.cvtColor(source1, cv2.COLOR_BGR2RGB)\n",
    "source2 = cv2.imread(\"./5.png\")\n",
    "source2 = cv2.cvtColor(source2, cv2.COLOR_BGR2RGB)\n",
    "prompt = 'best qualit'\n",
    "a_prompt = ''\n",
    "n_prompt = 'longbody, lowres, bad anatomy, bad hands, missing fingers, pubic hair,extra digit, fewer digits, cropped, worst quality, low quality'\n",
    "out = style_transfer_fixing_structure(det='Depth_Zoe',input_image1=source2,\n",
    "                                time_frac=0.3,detect_image = source1,noisylatend=None,\n",
    "                                ddim_sampler=ddim_sampler_depth, model= model_depth, a_prompt=a_prompt,\n",
    "                                n_prompt = n_prompt, num_samples = 1, \n",
    "                                image_resolution = 512, prompt=prompt, detect_resolution=512, ddim_steps = 100, guess_mode =None, strength = 1, scale = 5, seed = 4, eta = 1, low_threshold = 100, high_threshold = 200 )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from PIL import Image\n",
    "imgg = Image.fromarray(out[0])\n",
    "imgg "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from PIL import Image\n",
    "imgg = Image.fromarray(out[0])\n",
    "imgg.save(f'./Result/style_transfer/great_wall/0.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import imageio\n",
    "with imageio.get_writer(uri='test6.gif', mode='I', fps=8) as writer:\n",
    "    for i in range(30):\n",
    "        writer.append_data(imageio.imread(f'Result/comb_1/{i:03d}.png'))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import shutil\n",
    "import os\n",
    "from tqdm import trange\n",
    "from PIL import Image\n",
    "import yaml\n",
    "def interpolate_naive(model,img1, img2, num_frames, out_dir='blend'):\n",
    "    if isinstance(img1, Image.Image):\n",
    "        img1 = torch.tensor(np.array(img1)).permute(2,0,1).unsqueeze(0).cuda()\n",
    "        img2 = torch.tensor(np.array(img2)).permute(2,0,1).unsqueeze(0).cuda()\n",
    "    ldm = model\n",
    "    L1 = ldm.get_first_stage_encoding(ldm.encode_first_stage(img1.float() / 127.5 - 1.0))\n",
    "    L2 = ldm.get_first_stage_encoding(ldm.encode_first_stage(img2.float() / 127.5 - 1.0))\n",
    "    shutil.rmtree(out_dir, ignore_errors=True)\n",
    "    os.makedirs(out_dir)\n",
    "    for frame_ix in trange(1,num_frames-1):\n",
    "        frac = frame_ix/(num_frames-1)\n",
    "        latent = slerp(L1, L2, frac)\n",
    "        x_samples = ldm.decode_first_stage(latent).permute(0, 2, 3, 1)\n",
    "        x_samples = (x_samples * 127.5 + 127.5).cpu().numpy().clip(0, 255).astype(np.uint8)\n",
    "        Image.fromarray(x_samples[0]).save(f'{out_dir}/{frame_ix:03d}.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "source1 = Image.open(\"./1.png\")\n",
    "source2 = Image.open(\"./0.png\")\n",
    "interpolate_naive(model_depth,source1, source2, 10, out_dir='blend')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_latent_stack(model, img1, img2, timesteps, share_noise=True):\n",
    "    ldm = model\n",
    "    latents1 = [ldm.get_first_stage_encoding(ldm.encode_first_stage(img1))]\n",
    "    latents2 = [ldm.get_first_stage_encoding(ldm.encode_first_stage(img2))]\n",
    "    \n",
    "    t_prev = None\n",
    "    for t_now in timesteps[1:]:\n",
    "        noise = torch.randn_like(latents1[-1])\n",
    "        latents1.append(add_more_noise(model,latents1[-1], noise, t_now, t_prev))\n",
    "        if not share_noise:\n",
    "            noise = torch.randn_like(latents2[-1])\n",
    "        latents2.append(add_more_noise(model,latents2[-1], noise, t_now, t_prev))\n",
    "        t_prev = t_now\n",
    "    return latents1, latents2\n",
    "\n",
    "def add_more_noise(model, latents, noise, t2, t1=None):\n",
    "    ldm = model\n",
    "    if t1 is None:\n",
    "        return ldm.sqrt_alphas_cumprod[t2] * latents + \\\n",
    "            ldm.sqrt_one_minus_alphas_cumprod[t2] * noise\n",
    "\n",
    "    a1 = ldm.sqrt_alphas_cumprod[t1]\n",
    "    sig1 = ldm.sqrt_one_minus_alphas_cumprod[t1]\n",
    "    a2 = ldm.sqrt_alphas_cumprod[t2]\n",
    "    sig2 = ldm.sqrt_one_minus_alphas_cumprod[t2]\n",
    "\n",
    "    scale = a2/a1\n",
    "    sigma = (sig2**2 - (scale * sig1)**2).sqrt()\n",
    "    return scale * latents + sigma * noise\n",
    "def get_step_schedule(min_steps, max_steps, num_levels, schedule_type='convex'):\n",
    "    diff = max_steps - min_steps\n",
    "    if schedule_type == 'concave':\n",
    "        return [0]+[int(diff * x**.5)+min_steps for x in np.linspace(0, 1, num_levels)]\n",
    "    elif schedule_type == 'convex':\n",
    "        return [0]+[int(diff * x**2)+min_steps for x in np.linspace(0, 1, num_levels)]\n",
    "    elif schedule_type == 'linear':\n",
    "        return [0]+[int(x) for x in np.linspace(min_steps, max_steps, num_levels)]\n",
    "def interpolate_then_diffuse(img1, img2, input_image1, input_image2, model, num_frames,ddim_sampler, controls=None, control_type='pose', min_steps=.25, max_steps=.5, prompt=None, n_prompt=None, ddim_steps=250, guide_scale=7.5, schedule_type='linear', optimize_cond=0, cond_path=None, cond_lr=1e-4, out_dir='blend'): #steps_per_frame=10, \n",
    "    \"\"\"\n",
    "    each successive frame has more noise than the previous\n",
    "    \"\"\"\n",
    "    if min_steps < 1:\n",
    "        min_steps = int(ddim_steps * min_steps)\n",
    "    if max_steps < 1:\n",
    "        max_steps = int(ddim_steps * max_steps)\n",
    "    shutil.rmtree(out_dir, ignore_errors=True)\n",
    "    os.makedirs(out_dir)\n",
    "\n",
    "    if isinstance(img1, Image.Image):\n",
    "        img1.save(f'{out_dir}/{0:03d}.png')\n",
    "        img2.save(f'{out_dir}/{num_frames-1:03d}.png')\n",
    "        img1 = torch.tensor(np.array(img1)).permute(2,0,1).unsqueeze(0).cuda()\n",
    "        img2 = torch.tensor(np.array(img2)).permute(2,0,1).unsqueeze(0).cuda()\n",
    "    preprocessor = ZoeDetector()\n",
    "    ldm = model\n",
    "    ldm.control_scales = [1] * 13\n",
    "\n",
    "    if cond_path and os.path.exists(cond_path):\n",
    "        assert optimize_cond > 0\n",
    "        cond1, cond2, uncond_base = torch.load(cond_path)\n",
    "    else:\n",
    "        cond1 = ldm.get_learned_conditioning([prompt])\n",
    "        uncond_base = ldm.get_learned_conditioning([n_prompt])\n",
    "\n",
    "    cond = {\"c_crossattn\": [cond1], 'c_concat': None}\n",
    "    un_cond = {\"c_crossattn\": [uncond_base], 'c_concat': None}\n",
    "\n",
    "    img1 = img1.float() / 127.5 - 1.0\n",
    "    img2 = img2.float() / 127.5 - 1.0\n",
    "    # schedules include endpoints\n",
    "    ddim_sampler.make_schedule(ddim_steps, verbose=False)\n",
    "    step_schedule = get_step_schedule(min_steps, max_steps, (num_frames+1)//2, schedule_type=schedule_type)\n",
    "    timestep_schedule = [ddim_sampler.ddim_timesteps[s] for s in step_schedule]\n",
    "    latents1, latents2 = get_latent_stack(model,img1, img2, timestep_schedule)\n",
    "    latents = [None] * num_frames\n",
    "    latents[0] = latents1[0]\n",
    "    latents[-1] = latents2[0]\n",
    "    shape = latents[0].shape[-3:]\n",
    "    input_image1 = HWC3(input_image1)\n",
    "    input_image2 = HWC3(input_image2)\n",
    "\n",
    "    kwargs = dict(cond_lr=cond_lr, cond_steps=optimize_cond, prompt=prompt, n_prompt=n_prompt, ddim_steps=ddim_steps, guide_scale=guide_scale, step_schedule=step_schedule)\n",
    "    yaml.dump(kwargs, open(f'{out_dir}/args.yaml', 'w'))\n",
    "\n",
    "    for frame_ix in trange(1,num_frames-1):\n",
    "        frac = frame_ix/(num_frames-1)\n",
    "        f = min(frame_ix, num_frames - frame_ix - 1)\n",
    "        latents[frame_ix] = slerp(latents1[f], latents2[f], frac)\n",
    "        \n",
    "       \n",
    "        detected_map1 = preprocessor(resize_image(input_image1, detect_resolution))\n",
    "        detected_map2 = preprocessor(resize_image(input_image2, detect_resolution))\n",
    "        detected_map1 = HWC3(detected_map1)\n",
    "        detected_map2 = HWC3(detected_map2)\n",
    "        \n",
    "        img1 = resize_image(input_image1, image_resolution)\n",
    "        img2 = resize_image(input_image2, image_resolution)\n",
    "        H, W, C = img1.shape\n",
    "        detected_map1 = cv2.resize(detected_map1, (W, H), interpolation=cv2.INTER_LINEAR)\n",
    "        detected_map2 = cv2.resize(detected_map2, (W, H), interpolation=cv2.INTER_LINEAR)\n",
    "\n",
    "        control1 = torch.from_numpy(detected_map1.copy()).float().cuda() / 255.0\n",
    "        control2 = torch.from_numpy(detected_map2.copy()).float().cuda() / 255.0\n",
    "        control = slerp(control1,control2,frac)\n",
    "        control = torch.stack([control for _ in range(num_samples)], dim=0)\n",
    "        control = einops.rearrange(control, 'b h w c -> b c h w').clone()\n",
    "        cond[\"c_concat\"] = un_cond[\"c_concat\"] = [control]\n",
    "        \n",
    "        seed = random.randint(0, 65535)\n",
    "        seed_everything(seed)\n",
    "\n",
    "\n",
    "        cond = {\"c_concat\": [control], \"c_crossattn\": [model.get_learned_conditioning([prompt + ', ' + a_prompt] * num_samples)]}\n",
    "        un_cond = {\"c_concat\": None if guess_mode else [control], \"c_crossattn\": [model.get_learned_conditioning([n_prompt] * num_samples)]}\n",
    "        shape = (4, H // 8, W // 8)\n",
    "\n",
    "\n",
    "        samples, _ = ddim_sampler.sample(ddim_steps, 1,\n",
    "            shape, cond, verbose=False,\n",
    "            x_T=latents[frame_ix], timesteps=step_schedule[f],\n",
    "            unconditional_guidance_scale=guide_scale,eta = 1,\n",
    "            unconditional_conditioning=un_cond)\n",
    "\n",
    "        x_samples = ldm.decode_first_stage(samples).permute(0, 2, 3, 1)\n",
    "        x_samples = (x_samples * 127.5 + 127.5).cpu().numpy().clip(0, 255).astype(np.uint8)\n",
    "        Image.fromarray(x_samples[0]).save(f'{out_dir}/{frame_ix:03d}.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "source1 = Image.open(\"./1.png\")\n",
    "source2 = Image.open(\"./0.png\")\n",
    "input_one = cv2.imread(\"./1.png\")\n",
    "input_one = cv2.cvtColor(input_one, cv2.COLOR_BGR2RGB)\n",
    "input_two = cv2.imread(\"./0.png\")\n",
    "input_two = cv2.cvtColor(input_two, cv2.COLOR_BGR2RGB)\n",
    "\n",
    "interpolate_then_diffuse(img1 = source1, img2 = source2, input_image1 = input_one, input_image2 = input_two, model = model_depth, num_frames = 10,ddim_sampler = ddim_sampler_depth, controls=None, control_type='pose', \n",
    "                         min_steps=.25, max_steps=.5, prompt=prompt, n_prompt=n_prompt, ddim_steps=25, guide_scale=7.5, schedule_type='linear', optimize_cond=0, cond_path=None, cond_lr=1e-4, out_dir='blend')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from torchvision import transforms\n",
    "def learn_conditioning(model,ddim_sampler, img1, img2, cond_base, uncond_base, ddim_steps, guide_scale, num_iters=200, cond_lr=1e-4):\n",
    "        # augment = transforms.TrivialAugmentWide(num_magnitude_bins=20)\n",
    "        augment = transforms.Compose([\n",
    "            transforms.ColorJitter(brightness=0.1, contrast=0.2, saturation=0.2, hue=0.1),\n",
    "            transforms.RandomResizedCrop(size=(512,512), scale=(0.7,1.0)),\n",
    "        ])\n",
    "\n",
    "        cond = {\"c_crossattn\": [cond_base], 'c_concat': None}\n",
    "        un_cond = {\"c_crossattn\": [uncond_base], 'c_concat': None}\n",
    "        ldm = model\n",
    "        uncond_base.requires_grad_(True)\n",
    "        cond1 = cond_base\n",
    "        cond2 = cond_base.clone()\n",
    "        cond1.requires_grad_(True)\n",
    "        cond2.requires_grad_(True)\n",
    "        optimizer = torch.optim.Adam([cond1, cond2, uncond_base], lr=cond_lr) #\n",
    "        T = ddim_steps\n",
    "        ddim_sampler.make_schedule(T, verbose=False)\n",
    "        for cur_iter in range(num_iters):\n",
    "            L1 = ldm.get_first_stage_encoding(ldm.encode_first_stage(augment(img1).float() / 127.5 - 1.0))\n",
    "            L2 = ldm.get_first_stage_encoding(ldm.encode_first_stage(augment(img2).float() / 127.5 - 1.0))\n",
    "            with torch.autocast('cuda'):\n",
    "                u = np.random.randint(T//3, 2*T//3)\n",
    "                t_u = ddim_sampler.ddim_timesteps[u]\n",
    "                tu = torch.tensor([t_u], device='cuda', dtype=torch.long)\n",
    "\n",
    "                cond[\"c_crossattn\"] = [cond1]\n",
    "                noise = torch.randn_like(L1)\n",
    "                x_t_u = ldm.sqrt_alphas_cumprod[t_u] * L1 + \\\n",
    "                    ldm.sqrt_one_minus_alphas_cumprod[t_u] * noise\n",
    "                eps = ddim_sampler.pred_eps(x_t_u, cond, tu, unconditional_guidance_scale=guide_scale, unconditional_conditioning=un_cond)\n",
    "                loss1 = (eps - noise).pow(2).mean()\n",
    "                loss1.backward()\n",
    "                optimizer.step()\n",
    "                optimizer.zero_grad(set_to_none=True)\n",
    "\n",
    "                cond[\"c_crossattn\"] = [cond2]\n",
    "                noise = torch.randn_like(L2)\n",
    "                x_t_u = ldm.sqrt_alphas_cumprod[t_u] * L2 + \\\n",
    "                    ldm.sqrt_one_minus_alphas_cumprod[t_u] * noise\n",
    "                eps = ddim_sampler.pred_eps(x_t_u, cond, tu, unconditional_guidance_scale=guide_scale, unconditional_conditioning=un_cond)\n",
    "                loss2 = (eps - noise).pow(2).mean()\n",
    "                loss2.backward()\n",
    "\n",
    "                optimizer.step()\n",
    "                optimizer.zero_grad(set_to_none=True)\n",
    "                # if cur_iter % 50 == 0:\n",
    "                #     print(f'iter {cur_iter}: {loss1.item()}, {loss2.item()}')\n",
    "\n",
    "        cond1.requires_grad_(False)\n",
    "        cond2.requires_grad_(False)\n",
    "        uncond_base.requires_grad_(False)\n",
    "        return cond1, cond2, uncond_base\n",
    "def interpolate(model, img1, img2, input_image1, input_image2,ddim_sampler, controls=None, control_type='pose', prompt=None, n_prompt=None, min_steps=.25, \n",
    "                max_steps=.5, ddim_steps=250, num_frames=17, guide_scale=7.5, schedule_type='linear', optimize_cond=0, \n",
    "                cond_interp='spherical', cond_path=None, cond_lr=1e-4, bias=0, retroactive_interp=True, share_noise=True, out_dir='blend'):\n",
    "        \"\"\"\n",
    "        ddim_steps: number of steps in DDIM sampling\n",
    "        num_frames: includes endpoints (both original images)\n",
    "        \"\"\"\n",
    "        if min_steps < 1:\n",
    "            min_steps = int(ddim_steps * min_steps)\n",
    "        if max_steps < 1:\n",
    "            max_steps = int(ddim_steps * max_steps)\n",
    "        shutil.rmtree(out_dir, ignore_errors=True)\n",
    "        os.makedirs(out_dir)\n",
    "    \n",
    "        if isinstance(img1, Image.Image):\n",
    "            img1.save(f'{out_dir}/{0:03d}.png')\n",
    "            img2.save(f'{out_dir}/{num_frames-1:03d}.png')\n",
    "            img1 = torch.tensor(np.array(img1)).permute(2,0,1).unsqueeze(0).cuda()\n",
    "            img2 = torch.tensor(np.array(img2)).permute(2,0,1).unsqueeze(0).cuda()\n",
    "\n",
    "        preprocessor = ZoeDetector()\n",
    "        ldm = model\n",
    "        ldm.control_scales = [1] * 13\n",
    "\n",
    "        cond1 = ldm.get_learned_conditioning([prompt])\n",
    "        uncond_base = ldm.get_learned_conditioning([n_prompt])\n",
    "        if optimize_cond:\n",
    "            cond1, cond2, uncond_base = learn_conditioning( model=model, ddim_sampler=ddim_sampler,img1 =img1, img2 = img2, cond_base = cond1, uncond_base = uncond_base, ddim_steps = ddim_steps, guide_scale=guide_scale, num_iters=optimize_cond, cond_lr=cond_lr)\n",
    "        detected_map1 = preprocessor(resize_image(input_image1, detect_resolution))\n",
    "        detected_map2 = preprocessor(resize_image(input_image2, detect_resolution))\n",
    "        detected_map1 = HWC3(detected_map1)\n",
    "        detected_map2 = HWC3(detected_map2)\n",
    "        \n",
    "        img11 = resize_image(input_image1, image_resolution)\n",
    "        img21 = resize_image(input_image2, image_resolution)\n",
    "        H, W, C = img11.shape\n",
    "        detected_map1 = cv2.resize(detected_map1, (W, H), interpolation=cv2.INTER_LINEAR)\n",
    "        detected_map2 = cv2.resize(detected_map2, (W, H), interpolation=cv2.INTER_LINEAR)\n",
    "\n",
    "        control1 = torch.from_numpy(detected_map1.copy()).float().cuda() / 255.0\n",
    "        control2 = torch.from_numpy(detected_map2.copy()).float().cuda() / 255.0\n",
    "\n",
    "        cond = {\"c_crossattn\": [cond1], 'c_concat': None}\n",
    "        un_cond = {\"c_crossattn\": [uncond_base], 'c_concat': None}\n",
    "\n",
    "        img1 = img1.float() / 127.5 - 1.0\n",
    "        img2 = img2.float() / 127.5 - 1.0\n",
    "        # schedules include endpoints\n",
    "        num_levels = int(np.log2(num_frames-1)) # does not include endpoints\n",
    "        assert np.log2(num_frames-1) % 1 < 1e-5\n",
    "        ddim_sampler.make_schedule(ddim_steps, verbose=False)\n",
    "        step_schedule = get_step_schedule(min_steps, max_steps, num_levels, schedule_type=schedule_type)\n",
    "        timesteps = ddim_sampler.ddim_timesteps\n",
    "        timestep_schedule = [timesteps[s] for s in step_schedule]\n",
    "        latents1, latents2 = get_latent_stack(model,img1, img2, timestep_schedule, share_noise=share_noise)\n",
    "        latents = [None] * num_frames\n",
    "        latents[0] = latents1[0]\n",
    "        latents[-1] = latents2[0]\n",
    "        \n",
    "        kwargs = dict(cond_lr=cond_lr, cond_steps=optimize_cond, prompt=prompt, n_prompt=n_prompt, ddim_steps=ddim_steps, guide_scale=guide_scale, step_schedule=step_schedule, bias=bias, retroactive_interp=retroactive_interp, share_noise=share_noise)\n",
    "        yaml.dump(kwargs, open(f'{out_dir}/args.yaml', 'w'))\n",
    "        \n",
    "        interpolate_latents = slerp\n",
    "\n",
    "        for level in trange(1,num_levels+1):\n",
    "            cur_ix = step_schedule[-level]\n",
    "            prev_ix = step_schedule[-level-1]\n",
    "            latents[0] = latents1[-level]\n",
    "            latents[-1] = latents2[-level]\n",
    "            df = 2**(num_levels-level)\n",
    "\n",
    "            for frame_ix in range(df, num_frames-1, df*2):\n",
    "                frac = .5\n",
    "                if frame_ix-df == 0:\n",
    "                    frac -= bias\n",
    "                if frame_ix+df == num_frames-1:\n",
    "                    frac += bias\n",
    "                latents[frame_ix] = interpolate_latents(latents[frame_ix-df], latents[frame_ix+df], frac)\n",
    "\n",
    "            if retroactive_interp:\n",
    "                if level == 2:\n",
    "                    latents[num_frames//2] = interpolate_latents(latents[num_frames//4], latents[3*num_frames//4], .5)\n",
    "                \n",
    "                if level == 3:\n",
    "                    latents[num_frames//4] = interpolate_latents(latents[num_frames//8], latents[3*num_frames//8], .5)\n",
    "                    latents[num_frames//2] = interpolate_latents(latents[3*num_frames//8], latents[5*num_frames//8], .5)\n",
    "                    latents[3*num_frames//4] = interpolate_latents(latents[5*num_frames//8], latents[7*num_frames//8], .5)\n",
    "            \n",
    "            for frame_ix in range(df, num_frames-1, df): # exclude endpoints\n",
    "                frac = frame_ix/(num_frames-1)\n",
    "\n",
    "                control = slerp(control1,control2,frac)\n",
    "                control = torch.stack([control for _ in range(num_samples)], dim=0)\n",
    "                control = einops.rearrange(control, 'b h w c -> b c h w').clone()\n",
    "                cond[\"c_concat\"] = un_cond[\"c_concat\"] = [control]\n",
    "\n",
    "                \n",
    "                for i, t in enumerate(np.flip(timesteps[prev_ix:cur_ix])):\n",
    "                    index = cur_ix - i - 1\n",
    "                    ts = torch.tensor([t], device='cuda', dtype=torch.long)\n",
    " \n",
    "                    latents[frame_ix] = ddim_sampler.p_sample_ddim(latents[frame_ix], cond, ts, index=index, unconditional_guidance_scale=guide_scale,\n",
    "                    unconditional_conditioning=un_cond)[0]\n",
    "\n",
    "        for frame_ix in range(1,num_frames-1):\n",
    "            x_samples = ldm.decode_first_stage(latents[frame_ix]).permute(0, 2, 3, 1)\n",
    "            x_samples = (x_samples * 127.5 + 127.5).cpu().numpy().clip(0, 255).astype(np.uint8)\n",
    "            Image.fromarray(x_samples[0]).save(f'{out_dir}/{frame_ix:03d}.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    " \n",
    "from PIL import Image\n",
    " \n",
    "def image_contact(input_path,output_path,axis):\n",
    "    paths = input_path\n",
    "    img_array = ''\n",
    "    img = ''\n",
    "    for i, v in enumerate(paths):\n",
    "        if i == 0:\n",
    "            img = Image.open(v)  # 打开图片\n",
    "            img_array = np.array(img)  # 转化为np array对象\n",
    "        if i > 0:\n",
    "            img_array2 = np.array(Image.open(v))\n",
    "            img_array = np.concatenate((img_array, img_array2), axis=axis)  # 横向拼接\n",
    "            # img_array = np.concatenate((img_array, img_array2), axis=0)  # 纵向拼接\n",
    "            img = Image.fromarray(img_array)\n",
    "    img.save(output_path)\n",
    "def Return_compact_numpy(input_path,axis):\n",
    "    paths = input_path\n",
    "    img_array = ''\n",
    "    img = ''\n",
    "    for i, v in enumerate(paths):\n",
    "        if i == 0:\n",
    "            img = Image.open(v)  # 打开图片\n",
    "            img_array = np.array(img)  # 转化为np array对象\n",
    "        if i > 0:\n",
    "            img_array2 = np.array(Image.open(v))\n",
    "            img_array = np.concatenate((img_array, img_array2), axis=axis)  # 横向拼接\n",
    "            # img_array = np.concatenate((img_array, img_array2), axis=0)  # 纵向拼接\n",
    "    return img_array\n",
    "tep = []\n",
    "path_style = \"./Result/comb_18\"\n",
    "for i,filename in enumerate(os.listdir(path_style)):\n",
    "    tep.append(os.path.join(path_style,filename))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "row = 8\n",
    "columne = 10\n",
    "for i in range(row):\n",
    "    if i == 0:\n",
    "        img_array = Return_compact_numpy(tep[i*columne: (i+1)*columne],1)\n",
    "    else:\n",
    "        img_array_2 = Return_compact_numpy(tep[i*columne: (i+1)*columne],1)\n",
    "        img_array = np.concatenate((img_array, img_array_2), axis=0)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "img = Image.fromarray(img_array)\n",
    "img.save(\"./Result/comb_18/final.png\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    " \n",
    "from PIL import Image\n",
    " \n",
    "paths = ['./0.png' ,'./test_imgs/111.png','./test_imgs/001.png','./test_imgs/002.png','./test_imgs/003.png','./test_imgs/004.png']\n",
    "img_array = ''\n",
    "img = ''\n",
    "for i, v in enumerate(paths):\n",
    "    if i == 0:\n",
    "        img_array = resize_image_to_squra(v,512)\n",
    "\n",
    "        img = Image.open(v)  # 打开图片\n",
    "        img_array = np.array(img)  # 转化为np array对象\n",
    "    if i > 0:\n",
    "        img_array2 = resize_image_to_squra(v,512)\n",
    "\n",
    "        img_array = np.concatenate((img_array, img_array2), axis=1)  # 横向拼接\n",
    "        # img_array = np.concatenate((img_array, img_array2), axis=0)  # 纵向拼接\n",
    "        img = Image.fromarray(img_array)\n",
    " \n",
    " \n",
    " \n",
    "# 保存图片\n",
    "img.save('test1.jpg')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    " \n",
    "from PIL import Image\n",
    " \n",
    "paths = ['./style1.png','./style2.png']\n",
    "img_array = ''\n",
    "img = ''\n",
    "for i, v in enumerate(paths):\n",
    "    if i == 0:\n",
    "        img = Image.open(v)  # 打开图片\n",
    "        img_array = np.array(img)  # 转化为np array对象\n",
    "    if i > 0:\n",
    "        img_array2 = np.array(Image.open(v))\n",
    "        img_array = np.concatenate((img_array, img_array2), axis=0)  # 横向拼接\n",
    "        # img_array = np.concatenate((img_array, img_array2), axis=0)  # 纵向拼接\n",
    "        img = Image.fromarray(img_array)\n",
    " \n",
    " \n",
    " \n",
    " \n",
    " \n",
    "# 保存图片\n",
    "img.save('testt.jpg')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": 4
}