mirror of
https://github.com/invoke-ai/InvokeAI
synced 2024-08-30 20:32:17 +00:00
fix: Black linting
This commit is contained in:
blessedcoolant
@ -52,17 +52,17 @@
|
||||
"name": "stdout",
|
||||
"text": [
|
||||
"Cloning into 'latent-diffusion'...\n",
|
||||
"remote: Enumerating objects: 992, done.\u001B[K\n",
|
||||
"remote: Counting objects: 100% (695/695), done.\u001B[K\n",
|
||||
"remote: Compressing objects: 100% (397/397), done.\u001B[K\n",
|
||||
"remote: Total 992 (delta 375), reused 564 (delta 253), pack-reused 297\u001B[K\n",
|
||||
"remote: Enumerating objects: 992, done.\u001b[K\n",
|
||||
"remote: Counting objects: 100% (695/695), done.\u001b[K\n",
|
||||
"remote: Compressing objects: 100% (397/397), done.\u001b[K\n",
|
||||
"remote: Total 992 (delta 375), reused 564 (delta 253), pack-reused 297\u001b[K\n",
|
||||
"Receiving objects: 100% (992/992), 30.78 MiB | 29.43 MiB/s, done.\n",
|
||||
"Resolving deltas: 100% (510/510), done.\n",
|
||||
"Cloning into 'taming-transformers'...\n",
|
||||
"remote: Enumerating objects: 1335, done.\u001B[K\n",
|
||||
"remote: Counting objects: 100% (525/525), done.\u001B[K\n",
|
||||
"remote: Compressing objects: 100% (493/493), done.\u001B[K\n",
|
||||
"remote: Total 1335 (delta 58), reused 481 (delta 30), pack-reused 810\u001B[K\n",
|
||||
"remote: Enumerating objects: 1335, done.\u001b[K\n",
|
||||
"remote: Counting objects: 100% (525/525), done.\u001b[K\n",
|
||||
"remote: Compressing objects: 100% (493/493), done.\u001b[K\n",
|
||||
"remote: Total 1335 (delta 58), reused 481 (delta 30), pack-reused 810\u001b[K\n",
|
||||
"Receiving objects: 100% (1335/1335), 412.35 MiB | 30.53 MiB/s, done.\n",
|
||||
"Resolving deltas: 100% (267/267), done.\n",
|
||||
"Obtaining file:///content/taming-transformers\n",
|
||||
@ -73,23 +73,24 @@
|
||||
"Installing collected packages: taming-transformers\n",
|
||||
" Running setup.py develop for taming-transformers\n",
|
||||
"Successfully installed taming-transformers-0.0.1\n",
|
||||
"\u001B[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n",
|
||||
"\u001b[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n",
|
||||
"tensorflow 2.8.0 requires tf-estimator-nightly==2.8.0.dev2021122109, which is not installed.\n",
|
||||
"arviz 0.11.4 requires typing-extensions<4,>=3.7.4.3, but you have typing-extensions 4.1.1 which is incompatible.\u001B[0m\n"
|
||||
"arviz 0.11.4 requires typing-extensions<4,>=3.7.4.3, but you have typing-extensions 4.1.1 which is incompatible.\u001b[0m\n"
|
||||
]
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"#@title Installation\n",
|
||||
"# @title Installation\n",
|
||||
"!git clone https://github.com/CompVis/latent-diffusion.git\n",
|
||||
"!git clone https://github.com/CompVis/taming-transformers\n",
|
||||
"!pip install -e ./taming-transformers\n",
|
||||
"!pip install omegaconf>=2.0.0 pytorch-lightning>=1.0.8 torch-fidelity einops\n",
|
||||
"\n",
|
||||
"import sys\n",
|
||||
"\n",
|
||||
"sys.path.append(\".\")\n",
|
||||
"sys.path.append('./taming-transformers')\n",
|
||||
"from taming.models import vqgan "
|
||||
"sys.path.append(\"./taming-transformers\")\n",
|
||||
"from taming.models import vqgan"
|
||||
]
|
||||
},
|
||||
{
|
||||
@ -104,11 +105,11 @@
|
||||
{
|
||||
"cell_type": "code",
|
||||
"source": [
|
||||
"#@title Download\n",
|
||||
"%cd latent-diffusion/ \n",
|
||||
"# @title Download\n",
|
||||
"%cd latent-diffusion/\n",
|
||||
"\n",
|
||||
"!mkdir -p models/ldm/cin256-v2/\n",
|
||||
"!wget -O models/ldm/cin256-v2/model.ckpt https://ommer-lab.com/files/latent-diffusion/nitro/cin/model.ckpt "
|
||||
"!wget -O models/ldm/cin256-v2/model.ckpt https://ommer-lab.com/files/latent-diffusion/nitro/cin/model.ckpt"
|
||||
],
|
||||
"metadata": {
|
||||
"colab": {
|
||||
@ -203,7 +204,7 @@
|
||||
{
|
||||
"cell_type": "code",
|
||||
"source": [
|
||||
"#@title loading utils\n",
|
||||
"# @title loading utils\n",
|
||||
"import torch\n",
|
||||
"from omegaconf import OmegaConf\n",
|
||||
"\n",
|
||||
@ -212,7 +213,7 @@
|
||||
"\n",
|
||||
"def load_model_from_config(config, ckpt):\n",
|
||||
" print(f\"Loading model from {ckpt}\")\n",
|
||||
" pl_sd = torch.load(ckpt)#, map_location=\"cpu\")\n",
|
||||
" pl_sd = torch.load(ckpt) # , map_location=\"cpu\")\n",
|
||||
" sd = pl_sd[\"state_dict\"]\n",
|
||||
" model = instantiate_from_config(config.model)\n",
|
||||
" m, u = model.load_state_dict(sd, strict=False)\n",
|
||||
@ -222,7 +223,7 @@
|
||||
"\n",
|
||||
"\n",
|
||||
"def get_model():\n",
|
||||
" config = OmegaConf.load(\"configs/latent-diffusion/cin256-v2.yaml\") \n",
|
||||
" config = OmegaConf.load(\"configs/latent-diffusion/cin256-v2.yaml\")\n",
|
||||
" model = load_model_from_config(config, \"models/ldm/cin256-v2/model.ckpt\")\n",
|
||||
" return model"
|
||||
],
|
||||
@ -276,18 +277,18 @@
|
||||
{
|
||||
"cell_type": "code",
|
||||
"source": [
|
||||
"import numpy as np \n",
|
||||
"import numpy as np\n",
|
||||
"from PIL import Image\n",
|
||||
"from einops import rearrange\n",
|
||||
"from torchvision.utils import make_grid\n",
|
||||
"\n",
|
||||
"\n",
|
||||
"classes = [25, 187, 448, 992] # define classes to be sampled here\n",
|
||||
"classes = [25, 187, 448, 992] # define classes to be sampled here\n",
|
||||
"n_samples_per_class = 6\n",
|
||||
"\n",
|
||||
"ddim_steps = 20\n",
|
||||
"ddim_eta = 0.0\n",
|
||||
"scale = 3.0 # for unconditional guidance\n",
|
||||
"scale = 3.0 # for unconditional guidance\n",
|
||||
"\n",
|
||||
"\n",
|
||||
"all_samples = list()\n",
|
||||
@ -295,36 +296,39 @@
|
||||
"with torch.no_grad():\n",
|
||||
" with model.ema_scope():\n",
|
||||
" uc = model.get_learned_conditioning(\n",
|
||||
" {model.cond_stage_key: torch.tensor(n_samples_per_class*[1000]).to(model.device)}\n",
|
||||
" )\n",
|
||||
" \n",
|
||||
" {model.cond_stage_key: torch.tensor(n_samples_per_class * [1000]).to(model.device)}\n",
|
||||
" )\n",
|
||||
"\n",
|
||||
" for class_label in classes:\n",
|
||||
" print(f\"rendering {n_samples_per_class} examples of class '{class_label}' in {ddim_steps} steps and using s={scale:.2f}.\")\n",
|
||||
" xc = torch.tensor(n_samples_per_class*[class_label])\n",
|
||||
" print(\n",
|
||||
" f\"rendering {n_samples_per_class} examples of class '{class_label}' in {ddim_steps} steps and using s={scale:.2f}.\"\n",
|
||||
" )\n",
|
||||
" xc = torch.tensor(n_samples_per_class * [class_label])\n",
|
||||
" c = model.get_learned_conditioning({model.cond_stage_key: xc.to(model.device)})\n",
|
||||
" \n",
|
||||
" samples_ddim, _ = sampler.sample(S=ddim_steps,\n",
|
||||
" conditioning=c,\n",
|
||||
" batch_size=n_samples_per_class,\n",
|
||||
" shape=[3, 64, 64],\n",
|
||||
" verbose=False,\n",
|
||||
" unconditional_guidance_scale=scale,\n",
|
||||
" unconditional_conditioning=uc, \n",
|
||||
" eta=ddim_eta)\n",
|
||||
"\n",
|
||||
" samples_ddim, _ = sampler.sample(\n",
|
||||
" S=ddim_steps,\n",
|
||||
" conditioning=c,\n",
|
||||
" batch_size=n_samples_per_class,\n",
|
||||
" shape=[3, 64, 64],\n",
|
||||
" verbose=False,\n",
|
||||
" unconditional_guidance_scale=scale,\n",
|
||||
" unconditional_conditioning=uc,\n",
|
||||
" eta=ddim_eta,\n",
|
||||
" )\n",
|
||||
"\n",
|
||||
" x_samples_ddim = model.decode_first_stage(samples_ddim)\n",
|
||||
" x_samples_ddim = torch.clamp((x_samples_ddim+1.0)/2.0, \n",
|
||||
" min=0.0, max=1.0)\n",
|
||||
" x_samples_ddim = torch.clamp((x_samples_ddim + 1.0) / 2.0, min=0.0, max=1.0)\n",
|
||||
" all_samples.append(x_samples_ddim)\n",
|
||||
"\n",
|
||||
"\n",
|
||||
"# display as grid\n",
|
||||
"grid = torch.stack(all_samples, 0)\n",
|
||||
"grid = rearrange(grid, 'n b c h w -> (n b) c h w')\n",
|
||||
"grid = rearrange(grid, \"n b c h w -> (n b) c h w\")\n",
|
||||
"grid = make_grid(grid, nrow=n_samples_per_class)\n",
|
||||
"\n",
|
||||
"# to image\n",
|
||||
"grid = 255. * rearrange(grid, 'c h w -> h w c').cpu().numpy()\n",
|
||||
"grid = 255.0 * rearrange(grid, \"c h w -> h w c\").cpu().numpy()\n",
|
||||
"Image.fromarray(grid.astype(np.uint8))"
|
||||
],
|
||||
"metadata": {
|
||||
|
||||
Reference in New Issue
Block a user