Book a Demo!
CoCalc Logo Icon
StoreFeaturesDocsShareSupportNewsAboutPoliciesSign UpSign In
mfrashad
GitHub Repository: mfrashad/gancreate-saai
 Path: blob/main/text2art.ipynb
330 views
Kernel: Python 3 (ipykernel)
#@title New Setup
!wget https://www.dropbox.com/s/5t33zjcj29rnem/text2art.tar.gz
!tar -xzvf text2art.tar.gz
%cd clipit
!git config core.filemode false
!git pull --rebase
%cd ..
from IPython.utils import io
# with io.capture_output() as captured:
  # !pip install taming-transformers
!pip install ftfy regex tqdm omegaconf pytorch-lightning kornia imageio-ffmpeg    einops torch-optimizer easydict braceexpand
!pip install git+https://github.com/pvigier/perlin-numpy

# ClipDraw deps
!pip install svgwrite svgpathtools cssutils numba torch-tools visdom

!pip install gradio

!rm -rf diffvg
!git clone https://github.com/BachiLi/diffvg
%cd diffvg
!git submodule update --init --recursive
!python setup.py install
%cd ..

import sys
sys.path.append("clipit")
--2021-09-12 04:50:25-- https://www.dropbox.com/s/5t33zjcj29rnemp/text2art.tar.gz Resolving www.dropbox.com (www.dropbox.com)... 162.125.3.18, 2620:100:6018:18::a27d:312 Connecting to www.dropbox.com (www.dropbox.com)|162.125.3.18|:443... connected. HTTP request sent, awaiting response... 301 Moved Permanently Location: /s/raw/5t33zjcj29rnemp/text2art.tar.gz [following] --2021-09-12 04:50:25-- https://www.dropbox.com/s/raw/5t33zjcj29rnemp/text2art.tar.gz Reusing existing connection to www.dropbox.com:443. HTTP request sent, awaiting response... 302 Found Location: https://uc873a45d761514f888229a35ae9.dl.dropboxusercontent.com/cd/0/inline/BWBrkSAP8PidYCpiVr16awy-rOfW7lQA-24OiZekAGoFjz38eVM4XDTctGKeRlH2SW9AVsKImca2aIcD5GcqDkPaV1Ur3q5WmDZlOW-UgX2Lgpnqmo84mZ1fYXyFJhiEL4WqP1wCElm2xvjH9W6nYxTY/file# [following] --2021-09-12 04:50:26-- https://uc873a45d761514f888229a35ae9.dl.dropboxusercontent.com/cd/0/inline/BWBrkSAP8PidYCpiVr16awy-rOfW7lQA-24OiZekAGoFjz38eVM4XDTctGKeRlH2SW9AVsKImca2aIcD5GcqDkPaV1Ur3q5WmDZlOW-UgX2Lgpnqmo84mZ1fYXyFJhiEL4WqP1wCElm2xvjH9W6nYxTY/file Resolving uc873a45d761514f888229a35ae9.dl.dropboxusercontent.com (uc873a45d761514f888229a35ae9.dl.dropboxusercontent.com)... 162.125.3.15, 2620:100:6018:15::a27d:30f Connecting to uc873a45d761514f888229a35ae9.dl.dropboxusercontent.com (uc873a45d761514f888229a35ae9.dl.dropboxusercontent.com)|162.125.3.15|:443... connected. HTTP request sent, awaiting response... 302 Found Location: /cd/0/inline2/BWCD7lEudTm24Mcf-Ip8xhwa5DtDYii_TFpdjqJq01dR0a776ipSmn5fQqkZ0RNvvI_G2r7t5S3fNDwtEDyQoV2GaoacPsN9gugeLEe0LAXleQB_CcmNDI6xSg01mcQrJKdmd_KjObUenSy1kY7UXwEVEDaMp7xs0EKB2zLnH7pEPEo1m03ugwPonrWNytEAPYRYOKjzxpNLzWHONtnRT89fbq8qPxQeOYcJejj3Qu64IJ8bjTRjCVKZkiTpCIcrWECjcxjPcZhzV0MtKCzZa-1HUFv06rGXkEtaRWCRK9R4QlaQn_XI54bQ__jdXsMlN_tTG5O0Wp-c1D_tAeRUqJAZilScaO4u_ZzhZWwJ3HncYA0BdJaX9Lt3kNplgLKL7pc/file [following] --2021-09-12 04:50:26-- https://uc873a45d761514f888229a35ae9.dl.dropboxusercontent.com/cd/0/inline2/BWCD7lEudTm24Mcf-Ip8xhwa5DtDYii_TFpdjqJq01dR0a776ipSmn5fQqkZ0RNvvI_G2r7t5S3fNDwtEDyQoV2GaoacPsN9gugeLEe0LAXleQB_CcmNDI6xSg01mcQrJKdmd_KjObUenSy1kY7UXwEVEDaMp7xs0EKB2zLnH7pEPEo1m03ugwPonrWNytEAPYRYOKjzxpNLzWHONtnRT89fbq8qPxQeOYcJejj3Qu64IJ8bjTRjCVKZkiTpCIcrWECjcxjPcZhzV0MtKCzZa-1HUFv06rGXkEtaRWCRK9R4QlaQn_XI54bQ__jdXsMlN_tTG5O0Wp-c1D_tAeRUqJAZilScaO4u_ZzhZWwJ3HncYA0BdJaX9Lt3kNplgLKL7pc/file Reusing existing connection to uc873a45d761514f888229a35ae9.dl.dropboxusercontent.com:443. HTTP request sent, awaiting response... 200 OK Length: 2429560071 (2.3G) [application/octet-stream] Saving to: ‘text2art.tar.gz’ text2art.tar.gz 100%[===================>] 2.26G 76.1MB/s in 28s 2021-09-12 04:50:55 (82.5 MB/s) - ‘text2art.tar.gz’ saved [2429560071/2429560071] 
#@title Old Setup

from IPython.utils import io
with io.capture_output() as captured:
  !git clone https://github.com/openai/CLIP
  # !pip install taming-transformers
  !git clone https://github.com/CompVis/taming-transformers.git
  !rm -Rf clipit
  !git clone https://github.com/mfrashad/clipit.git
  !pip install ftfy regex tqdm omegaconf pytorch-lightning
  !pip install kornia
  !pip install imageio-ffmpeg   
  !pip install einops
  !pip install torch-optimizer
  !pip install easydict
  !pip install braceexpand
  !pip install git+https://github.com/pvigier/perlin-numpy

  # ClipDraw deps
  !pip install svgwrite
  !pip install svgpathtools
  !pip install cssutils
  !pip install numba
  !pip install torch-tools
  !pip install visdom

  !pip install gradio

  !git clone https://github.com/BachiLi/diffvg
  %cd diffvg
  # !ls
  !git submodule update --init --recursive
  !python setup.py install
  %cd ..

import sys
sys.path.append("clipit")


from importlib import reload  
reload(clipit)
<module 'clipit' from 'clipit/clipit.py'>
import clipit

#@title Text2Art Gradio UI
import gradio as gr
import os.path as osp
import platform
import argparse
import time
import sys
import subprocess
from IPython.display import HTML
from base64 import b64encode
import datetime
import torch
import clipit

import firebase_admin
from firebase_admin import credentials, firestore, storage

if not firebase_admin._apps:
    cred = credentials.Certificate("text2art-firebase-adminsdk-xig66-1abd6a59a5.json")
    firebase_admin.initialize_app(cred, {
        'storageBucket': 'text2art.appspot.com'
    })

db = firestore.client()
bucket = storage.bucket()

def generate(prompt, quality, style, aspect, email, public):
    torch.cuda.empty_cache()
    clipit.reset_settings()
    
    use_pixeldraw = (style == 'pixel art')
    use_clipdraw = (style == 'painting')
    seed = int(datetime.datetime.now().timestamp())
    clipit.add_settings(prompts=prompt,
                        seed=seed,
                        aspect=aspect,
                        iterations = 800 if (quality == 'best') else None,
                        quality= 'better' if (quality == 'best') else quality,
                        scale=2.5,
                        display_every=7919,
                        use_pixeldraw=use_pixeldraw,
                        use_clipdraw=use_clipdraw,
                        make_video=True)
    
    settings = clipit.apply_settings()
    clipit.do_init(settings)
    clipit.do_run(settings)

    doc_ref = db.collection('generated_images').add({})
    doc_id = doc_ref[1].id

    blob = bucket.blob(f'generated_images/{doc_id}.png')
    blob.upload_from_filename('output.png')
    blob.make_public()

    data = {
        "image": blob.public_url,
        "seed": seed,
        "prompt": prompt,
        "quality": quality,
        "aspect": aspect,
        "type": style,
        "user": email,
        "likes": 0,
        "public": public,
        "created_at": datetime.datetime.now()
    }
    db.collection('generated_images').document(doc_id).set(data)

    return 'output.png', 'output.mp4'


prompt = gr.inputs.Textbox(default="Underwater city", label="Text Prompt")
quality = gr.inputs.Radio(choices=['draft', 'normal', 'better', 'best'], label="Quality")
style = gr.inputs.Radio(choices=['image', 'painting','pixel art'], label="Type")
aspect = gr.inputs.Radio(choices=['square', 'widescreen','portrait'], label="Size")
email = gr.inputs.Textbox(placeholder="[email protected]", label="Email")
public = gr.inputs.Checkbox(default=True, label="Public")
# width = gr.inputs.Textbox(placeholder="width", label="Width")

inputs = []



iface = gr.Interface(generate, inputs=[prompt, quality, style, aspect, email, public], outputs=['image', 'video'], enable_queue=True, live=False)
iface.launch(debug=True)
Imageio: 'ffmpeg-linux64-v3.3.1' was not found on your computer; downloading it now. Try 1. Download from https://github.com/imageio/imageio-binaries/raw/master/ffmpeg/ffmpeg-linux64-v3.3.1 (43.8 MB) Downloading: 8192/45929032 bytes (0.03432448/45929032 bytes (7.5%7331840/45929032 bytes (16.011083776/45929032 bytes (24.1%14974976/45929032 bytes (32.6%19005440/45929032 bytes (41.4%23076864/45929032 bytes (50.2%27164672/45929032 bytes (59.1%31072256/45929032 bytes (67.7%35069952/45929032 bytes (76.4%39084032/45929032 bytes (85.1%43106304/45929032 bytes (93.9%45929032/45929032 bytes (100.0%) Done File saved as /root/.imageio/ffmpeg/ffmpeg-linux64-v3.3.1. Colab notebook detected. This cell will run indefinitely so that you can see errors and logs. To turn off, set debug=False in launch(). This share link will expire in 24 hours. If you need a permanent link, visit: https://gradio.app/introducing-hosted (NEW!) Running on External URL: https://33847.gradio.app Interface loading below...
[2021-09-01 17:29:03,791] ERROR in app: Exception on /api/predict/ [POST] Traceback (most recent call last): File "/usr/local/lib/python3.7/dist-packages/flask/app.py", line 2447, in wsgi_app response = self.full_dispatch_request() File "/usr/local/lib/python3.7/dist-packages/flask/app.py", line 1952, in full_dispatch_request rv = self.handle_user_exception(e) File "/usr/local/lib/python3.7/dist-packages/flask_cors/extension.py", line 165, in wrapped_function return cors_after_request(app.make_response(f(*args, **kwargs))) File "/usr/local/lib/python3.7/dist-packages/flask/app.py", line 1821, in handle_user_exception reraise(exc_type, exc_value, tb) File "/usr/local/lib/python3.7/dist-packages/flask/_compat.py", line 39, in reraise raise value File "/usr/local/lib/python3.7/dist-packages/flask/app.py", line 1950, in full_dispatch_request rv = self.dispatch_request() File "/usr/local/lib/python3.7/dist-packages/flask/app.py", line 1936, in dispatch_request return self.view_functions[rule.endpoint](**req.view_args) File "/usr/local/lib/python3.7/dist-packages/gradio/networking.py", line 92, in wrapper return func(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/gradio/networking.py", line 180, in predict prediction, durations = app.interface.process(raw_input) File "/usr/local/lib/python3.7/dist-packages/gradio/interface.py", line 333, in process predictions, durations = self.run_prediction(processed_input, return_duration=True) File "/usr/local/lib/python3.7/dist-packages/gradio/interface.py", line 306, in run_prediction prediction = predict_fn(*processed_input) File "<ipython-input-5-86f3a67ffc40>", line 46, in generate clipit.do_init(settings) File "clipit/clipit.py", line 375, in do_init drawer = PixelDrawer(args.size[0], args.size[1], args.do_mono, [40, 40], scale=args.pixel_scale) NameError: name 'PixelDrawer' is not defined [2021-09-01 17:29:26,852] ERROR in app: Exception on /api/predict/ [POST] Traceback (most recent call last): File "/usr/local/lib/python3.7/dist-packages/flask/app.py", line 2447, in wsgi_app response = self.full_dispatch_request() File "/usr/local/lib/python3.7/dist-packages/flask/app.py", line 1952, in full_dispatch_request rv = self.handle_user_exception(e) File "/usr/local/lib/python3.7/dist-packages/flask_cors/extension.py", line 165, in wrapped_function return cors_after_request(app.make_response(f(*args, **kwargs))) File "/usr/local/lib/python3.7/dist-packages/flask/app.py", line 1821, in handle_user_exception reraise(exc_type, exc_value, tb) File "/usr/local/lib/python3.7/dist-packages/flask/_compat.py", line 39, in reraise raise value File "/usr/local/lib/python3.7/dist-packages/flask/app.py", line 1950, in full_dispatch_request rv = self.dispatch_request() File "/usr/local/lib/python3.7/dist-packages/flask/app.py", line 1936, in dispatch_request return self.view_functions[rule.endpoint](**req.view_args) File "/usr/local/lib/python3.7/dist-packages/gradio/networking.py", line 92, in wrapper return func(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/gradio/networking.py", line 180, in predict prediction, durations = app.interface.process(raw_input) File "/usr/local/lib/python3.7/dist-packages/gradio/interface.py", line 333, in process predictions, durations = self.run_prediction(processed_input, return_duration=True) File "/usr/local/lib/python3.7/dist-packages/gradio/interface.py", line 306, in run_prediction prediction = predict_fn(*processed_input) File "<ipython-input-5-86f3a67ffc40>", line 46, in generate clipit.do_init(settings) File "clipit/clipit.py", line 377, in do_init drawer = PixelDrawer(args.size[0], args.size[1], args.do_mono, scale=args.pixel_scale) NameError: name 'PixelDrawer' is not defined
Working with z of shape (1, 256, 16, 16) = 65536 dimensions.
Downloading: "https://download.pytorch.org/models/vgg16-397923af.pth" to /root/.cache/torch/hub/checkpoints/vgg16-397923af.pth
0%| | 0.00/528M [00:00<?, ?B/s]
loaded pretrained LPIPS loss from taming/modules/autoencoder/lpips/vgg.pth VQLPIPSWithDiscriminator running with hinge loss. Restored from models/vqgan_imagenet_f16_16384.ckpt
100%|████████████████████████████████████████| 338M/338M [00:01<00:00, 191MiB/s] 100%|███████████████████████████████████████| 335M/335M [00:05<00:00, 58.8MiB/s]
Using device: cuda:0 Optimising using: AdamP Using text prompts: ['archer #pixelart'] Using seed: 1630517379
0it [00:00, ?it/s]
iter: 0, loss: 1.86574, losses: 0.941887, 0.923851
Image in a Jupyter notebook
iter: 10, loss: 1.64829, losses: 0.812154, 0.836135 iter: 20, loss: 1.5564, losses: 0.779102, 0.777302 iter: 30, loss: 1.513, losses: 0.762711, 0.75029 iter: 40, loss: 1.40023, losses: 0.713814, 0.686411 iter: 50, loss: 1.45874, losses: 0.737288, 0.721451 iter: 60, loss: 1.37977, losses: 0.698676, 0.681093 iter: 70, loss: 1.4185, losses: 0.721164, 0.697336 iter: 80, loss: 1.42715, losses: 0.719346, 0.7078 iter: 90, loss: 1.40306, losses: 0.71441, 0.688653 iter: 100, loss: 1.40595, losses: 0.710291, 0.695658 iter: 110, loss: 1.3659, losses: 0.693402, 0.672496 iter: 120, loss: 1.37344, losses: 0.697438, 0.675998 iter: 130, loss: 1.38128, losses: 0.69982, 0.681459 iter: 140, loss: 1.37968, losses: 0.69899, 0.680693 iter: 150, loss: 1.38226, losses: 0.698401, 0.683854 iter: 160, loss: 1.35614, losses: 0.691938, 0.664198 iter: 170, loss: 1.36098, losses: 0.690409, 0.670575 iter: 180, loss: 1.29059, losses: 0.660394, 0.630199 iter: 190, loss: 1.37913, losses: 0.699257, 0.679874 iter: 200, loss: 1.38251, losses: 0.700851, 0.68166 iter: 210, loss: 1.281, losses: 0.653468, 0.627529 iter: 220, loss: 1.27058, losses: 0.644167, 0.626413 iter: 230, loss: 1.37044, losses: 0.695182, 0.675262 iter: 240, loss: 1.37079, losses: 0.691429, 0.679364 iter: 250, loss: 1.36137, losses: 0.691814, 0.669559 iter: 260, loss: 1.27892, losses: 0.653048, 0.625876 iter: 270, loss: 1.27198, losses: 0.646997, 0.624983 iter: 280, loss: 1.33385, losses: 0.675659, 0.65819 iter: 290, loss: 1.32974, losses: 0.672196, 0.657546 iter: 300, loss: 1.25415, losses: 0.638933, 0.615217 iter: 310, loss: 1.24065, losses: 0.627161, 0.613488 iter: 320, loss: 1.33969, losses: 0.674016, 0.665671 iter: 330, loss: 1.37901, losses: 0.69734, 0.681673 iter: 340, loss: 1.3215, losses: 0.66868, 0.652819 iter: 350, loss: 1.22046, losses: 0.619954, 0.600506 [MoviePy] >>>> Building video output.mp4 [MoviePy] Writing video output.mp4
100%|██████████| 350/350 [00:07<00:00, 49.41it/s]
[MoviePy] Done. [MoviePy] >>>> Video ready: output.mp4 Working with z of shape (1, 256, 16, 16) = 65536 dimensions. loaded pretrained LPIPS loss from taming/modules/autoencoder/lpips/vgg.pth VQLPIPSWithDiscriminator running with hinge loss. Restored from models/vqgan_imagenet_f16_16384.ckpt Using device: cuda:0 Optimising using: AdamP Using text prompts: ['archer #pixelart'] Using seed: 1630517750
0it [00:00, ?it/s]
iter: 0, loss: 1.8718, losses: 0.947556, 0.924244
Image in a Jupyter notebook
iter: 10, loss: 1.71835, losses: 0.855928, 0.862425 iter: 20, loss: 1.6089, losses: 0.800597, 0.808306 iter: 30, loss: 1.52923, losses: 0.763806, 0.765421 iter: 40, loss: 1.42945, losses: 0.713453, 0.716 iter: 50, loss: 1.40652, losses: 0.714551, 0.691966 iter: 60, loss: 1.48862, losses: 0.749222, 0.739394 iter: 70, loss: 1.45489, losses: 0.732332, 0.722559 iter: 80, loss: 1.38517, losses: 0.700765, 0.684401 iter: 90, loss: 1.44498, losses: 0.726026, 0.718952 iter: 100, loss: 1.35689, losses: 0.68074, 0.676153 iter: 110, loss: 1.4336, losses: 0.71979, 0.71381 iter: 120, loss: 1.41403, losses: 0.711853, 0.702172 iter: 130, loss: 1.3478, losses: 0.682962, 0.664835 iter: 140, loss: 1.40981, losses: 0.708846, 0.700968 iter: 150, loss: 1.415, losses: 0.709339, 0.705663 iter: 160, loss: 1.39921, losses: 0.704246, 0.694966 iter: 170, loss: 1.38998, losses: 0.701806, 0.688171 iter: 180, loss: 1.31668, losses: 0.661921, 0.654759 iter: 190, loss: 1.32897, losses: 0.671421, 0.657554 iter: 200, loss: 1.40634, losses: 0.70739, 0.698952 iter: 210, loss: 1.39567, losses: 0.701002, 0.694667 iter: 220, loss: 1.38748, losses: 0.70105, 0.686433 iter: 230, loss: 1.31765, losses: 0.661579, 0.656075 iter: 240, loss: 1.39534, losses: 0.701635, 0.693703 iter: 250, loss: 1.30516, losses: 0.660302, 0.644857 iter: 260, loss: 1.38067, losses: 0.696629, 0.68404 iter: 270, loss: 1.37256, losses: 0.691031, 0.681533 iter: 280, loss: 1.29011, losses: 0.650374, 0.63974 iter: 290, loss: 1.38693, losses: 0.701415, 0.685517 iter: 300, loss: 1.3739, losses: 0.691449, 0.682449 iter: 310, loss: 1.34479, losses: 0.676744, 0.66805 iter: 320, loss: 1.36754, losses: 0.689587, 0.677951 iter: 330, loss: 1.36178, losses: 0.688559, 0.673225 iter: 340, loss: 1.36472, losses: 0.686502, 0.678217 iter: 350, loss: 1.3688, losses: 0.687334, 0.681465 [MoviePy] >>>> Building video output.mp4 [MoviePy] Writing video output.mp4
100%|██████████| 350/350 [00:07<00:00, 48.44it/s]
[MoviePy] Done. [MoviePy] >>>> Video ready: output.mp4
--------------------------------------------------------------------------- KeyboardInterrupt Traceback (most recent call last) <ipython-input-5-86f3a67ffc40> in <module>() 84 85 iface = gr.Interface(generate, inputs=[prompt, quality, style, aspect, email, public], outputs=['image', 'video'], enable_queue=True, live=False) ---> 86 iface.launch(debug=True) /usr/local/lib/python3.7/dist-packages/gradio/interface.py in launch(self, inline, inbrowser, share, debug, auth, auth_message, private_endpoint, prevent_thread_lock) 591 while True: 592 sys.stdout.flush() --> 593 time.sleep(0.1) 594 is_in_interactive_mode = bool(getattr(sys, 'ps1', sys.flags.interactive)) 595 if not prevent_thread_lock and not is_in_interactive_mode: KeyboardInterrupt:

Text2Art

!pip install fastapi nest-asyncio pyngrok uvicorn aiofiles python-multipart firebase-admin
!pip install sendgrid
Collecting fastapi Downloading fastapi-0.68.1-py3-none-any.whl (52 kB) |████████████████████████████████| 52 kB 761 kB/s eta 0:00:011 Requirement already satisfied: nest-asyncio in /usr/local/lib/python3.7/dist-packages (1.5.1) Collecting pyngrok Downloading pyngrok-5.1.0.tar.gz (745 kB) |████████████████████████████████| 745 kB 9.9 MB/s Collecting uvicorn Downloading uvicorn-0.15.0-py3-none-any.whl (54 kB) |████████████████████████████████| 54 kB 2.7 MB/s Collecting aiofiles Downloading aiofiles-0.7.0-py3-none-any.whl (13 kB) Collecting python-multipart Downloading python-multipart-0.0.5.tar.gz (32 kB) Requirement already satisfied: firebase-admin in /usr/local/lib/python3.7/dist-packages (4.4.0) Collecting starlette==0.14.2 Downloading starlette-0.14.2-py3-none-any.whl (60 kB) |████████████████████████████████| 60 kB 7.8 MB/s Collecting pydantic!=1.7,!=1.7.1,!=1.7.2,!=1.7.3,!=1.8,!=1.8.1,<2.0.0,>=1.6.2 Downloading pydantic-1.8.2-cp37-cp37m-manylinux2014_x86_64.whl (10.1 MB) |████████████████████████████████| 10.1 MB 29.6 MB/s Requirement already satisfied: typing-extensions>=3.7.4.3 in /usr/local/lib/python3.7/dist-packages (from pydantic!=1.7,!=1.7.1,!=1.7.2,!=1.7.3,!=1.8,!=1.8.1,<2.0.0,>=1.6.2->fastapi) (3.7.4.3) Requirement already satisfied: PyYAML in /usr/local/lib/python3.7/dist-packages (from pyngrok) (5.4.1) Collecting asgiref>=3.4.0 Downloading asgiref-3.4.1-py3-none-any.whl (25 kB) Requirement already satisfied: click>=7.0 in /usr/local/lib/python3.7/dist-packages (from uvicorn) (7.1.2) Collecting h11>=0.8 Downloading h11-0.12.0-py3-none-any.whl (54 kB) |████████████████████████████████| 54 kB 3.5 MB/s Requirement already satisfied: six>=1.4.0 in /usr/local/lib/python3.7/dist-packages (from python-multipart) (1.15.0) Requirement already satisfied: google-api-core[grpc]<2.0.0dev,>=1.14.0 in /usr/local/lib/python3.7/dist-packages (from firebase-admin) (1.26.3) Requirement already satisfied: cachecontrol>=0.12.6 in /usr/local/lib/python3.7/dist-packages (from firebase-admin) (0.12.6) Requirement already satisfied: google-api-python-client>=1.7.8 in /usr/local/lib/python3.7/dist-packages (from firebase-admin) (1.12.8) Requirement already satisfied: google-cloud-firestore>=1.4.0 in /usr/local/lib/python3.7/dist-packages (from firebase-admin) (1.7.0) Requirement already satisfied: google-cloud-storage>=1.18.0 in /usr/local/lib/python3.7/dist-packages (from firebase-admin) (1.18.1) Requirement already satisfied: requests in /usr/local/lib/python3.7/dist-packages (from cachecontrol>=0.12.6->firebase-admin) (2.23.0) Requirement already satisfied: msgpack>=0.5.2 in /usr/local/lib/python3.7/dist-packages (from cachecontrol>=0.12.6->firebase-admin) (1.0.2) Requirement already satisfied: protobuf>=3.12.0 in /usr/local/lib/python3.7/dist-packages (from google-api-core[grpc]<2.0.0dev,>=1.14.0->firebase-admin) (3.17.3) Requirement already satisfied: pytz in /usr/local/lib/python3.7/dist-packages (from google-api-core[grpc]<2.0.0dev,>=1.14.0->firebase-admin) (2018.9) Requirement already satisfied: packaging>=14.3 in /usr/local/lib/python3.7/dist-packages (from google-api-core[grpc]<2.0.0dev,>=1.14.0->firebase-admin) (21.0) Requirement already satisfied: google-auth<2.0dev,>=1.21.1 in /usr/local/lib/python3.7/dist-packages (from google-api-core[grpc]<2.0.0dev,>=1.14.0->firebase-admin) (1.34.0) Requirement already satisfied: googleapis-common-protos<2.0dev,>=1.6.0 in /usr/local/lib/python3.7/dist-packages (from google-api-core[grpc]<2.0.0dev,>=1.14.0->firebase-admin) (1.53.0) Requirement already satisfied: setuptools>=40.3.0 in /usr/local/lib/python3.7/dist-packages (from google-api-core[grpc]<2.0.0dev,>=1.14.0->firebase-admin) (57.4.0) Requirement already satisfied: grpcio<2.0dev,>=1.29.0 in /usr/local/lib/python3.7/dist-packages (from google-api-core[grpc]<2.0.0dev,>=1.14.0->firebase-admin) (1.39.0) Requirement already satisfied: google-auth-httplib2>=0.0.3 in /usr/local/lib/python3.7/dist-packages (from google-api-python-client>=1.7.8->firebase-admin) (0.0.4) Requirement already satisfied: httplib2<1dev,>=0.15.0 in /usr/local/lib/python3.7/dist-packages (from google-api-python-client>=1.7.8->firebase-admin) (0.17.4) Requirement already satisfied: uritemplate<4dev,>=3.0.0 in /usr/local/lib/python3.7/dist-packages (from google-api-python-client>=1.7.8->firebase-admin) (3.0.1) Requirement already satisfied: cachetools<5.0,>=2.0.0 in /usr/local/lib/python3.7/dist-packages (from google-auth<2.0dev,>=1.21.1->google-api-core[grpc]<2.0.0dev,>=1.14.0->firebase-admin) (4.2.2) Requirement already satisfied: pyasn1-modules>=0.2.1 in /usr/local/lib/python3.7/dist-packages (from google-auth<2.0dev,>=1.21.1->google-api-core[grpc]<2.0.0dev,>=1.14.0->firebase-admin) (0.2.8) Requirement already satisfied: rsa<5,>=3.1.4 in /usr/local/lib/python3.7/dist-packages (from google-auth<2.0dev,>=1.21.1->google-api-core[grpc]<2.0.0dev,>=1.14.0->firebase-admin) (4.7.2) Requirement already satisfied: google-cloud-core<2.0dev,>=1.0.3 in /usr/local/lib/python3.7/dist-packages (from google-cloud-firestore>=1.4.0->firebase-admin) (1.0.3) Requirement already satisfied: google-resumable-media<0.5.0dev,>=0.3.1 in /usr/local/lib/python3.7/dist-packages (from google-cloud-storage>=1.18.0->firebase-admin) (0.4.1) Requirement already satisfied: pyparsing>=2.0.2 in /usr/local/lib/python3.7/dist-packages (from packaging>=14.3->google-api-core[grpc]<2.0.0dev,>=1.14.0->firebase-admin) (2.4.7) Requirement already satisfied: pyasn1<0.5.0,>=0.4.6 in /usr/local/lib/python3.7/dist-packages (from pyasn1-modules>=0.2.1->google-auth<2.0dev,>=1.21.1->google-api-core[grpc]<2.0.0dev,>=1.14.0->firebase-admin) (0.4.8) Requirement already satisfied: idna<3,>=2.5 in /usr/local/lib/python3.7/dist-packages (from requests->cachecontrol>=0.12.6->firebase-admin) (2.10) Requirement already satisfied: chardet<4,>=3.0.2 in /usr/local/lib/python3.7/dist-packages (from requests->cachecontrol>=0.12.6->firebase-admin) (3.0.4) Requirement already satisfied: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in /usr/local/lib/python3.7/dist-packages (from requests->cachecontrol>=0.12.6->firebase-admin) (1.24.3) Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.7/dist-packages (from requests->cachecontrol>=0.12.6->firebase-admin) (2021.5.30) Building wheels for collected packages: pyngrok, python-multipart Building wheel for pyngrok (setup.py) ... done Created wheel for pyngrok: filename=pyngrok-5.1.0-py3-none-any.whl size=19006 sha256=5237c6093cf76e588347524662be61e5d5c386aa27ba28cb16688cd12fa0b1ea Stored in directory: /root/.cache/pip/wheels/bf/e6/af/ccf6598ecefecd44104069371795cb9b3afbcd16987f6ccfb3 Building wheel for python-multipart (setup.py) ... done Created wheel for python-multipart: filename=python_multipart-0.0.5-py3-none-any.whl size=31678 sha256=6381cfa040ed93edeac68ecfbad3679038fc23b4ba338d2340655c0be1eea0f2 Stored in directory: /root/.cache/pip/wheels/2c/41/7c/bfd1c180534ffdcc0972f78c5758f89881602175d48a8bcd2c Successfully built pyngrok python-multipart Installing collected packages: starlette, pydantic, h11, asgiref, uvicorn, python-multipart, pyngrok, fastapi, aiofiles Successfully installed aiofiles-0.7.0 asgiref-3.4.1 fastapi-0.68.1 h11-0.12.0 pydantic-1.8.2 pyngrok-5.1.0 python-multipart-0.0.5 starlette-0.14.2 uvicorn-0.15.0 Collecting sendgrid Downloading sendgrid-6.8.1-py3-none-any.whl (77 kB) |████████████████████████████████| 77 kB 3.1 MB/s Collecting python-http-client>=3.2.1 Downloading python_http_client-3.3.2.tar.gz (7.8 kB) Collecting starkbank-ecdsa>=1.0.0 Downloading starkbank-ecdsa-1.1.1.tar.gz (12 kB) Building wheels for collected packages: python-http-client, starkbank-ecdsa Building wheel for python-http-client (setup.py) ... done Created wheel for python-http-client: filename=python_http_client-3.3.2-py3-none-any.whl size=8365 sha256=d2e21baeb76fdc3409a99dd59be631203bd41aa6aa4a5df076efccfa28e15756 Stored in directory: /root/.cache/pip/wheels/91/f8/3f/61db5a99bc208c9f2d0ad95e343bd889a69ec98ec1b5c7005c Building wheel for starkbank-ecdsa (setup.py) ... done Created wheel for starkbank-ecdsa: filename=starkbank_ecdsa-1.1.1-py3-none-any.whl size=13653 sha256=e0e0eb29d0fbc1dc89b919aefc73f6d88699a49440298549baed7fd5eb06697c Stored in directory: /root/.cache/pip/wheels/c9/0d/2d/1630202621c9cdcae0d6548e0df73557f7197c2436a1e0b4a7 Successfully built python-http-client starkbank-ecdsa Installing collected packages: starkbank-ecdsa, python-http-client, sendgrid Successfully installed python-http-client-3.3.2 sendgrid-6.8.1 starkbank-ecdsa-1.1.1 
#@title MailGun function

import requests
def email_results_mailgun(email, prompt):
  
    return requests.post(
        "https://api.mailgun.net/v3/text2art.com/messages",
        auth=("api", "YOUR_API_KEY"),
        files=[("attachment",("output.png", open("output.png", "rb").read() )),
               ("attachment", ("output.mp4", open("output.mp4", "rb").read() ))],
        data={"from": "Text2Art <[email protected]>",
              "to": email,
              "subject": "Your Artwork is ready!",
              "text": f'Your generated arts using the prompt "{prompt}".',
              "html": f'Your generated arts using the prompt <strong>"{prompt}"</strong>.'})

#@title Watermark image

import random
from PIL import Image, ImageDraw, ImageFont

#Opening Image & Creating New Text Layer
def watermark_image(input_file, output_file):
    img = Image.open(input_file).convert("RGBA")
    txt = Image.new('RGBA', img.size, (255,255,255,0))

    #Creating Text
    text = "Text2Art.com"
    font = ImageFont.load_default()

    #Creating Draw Object
    d = ImageDraw.Draw(txt)

    #Positioning Text
    width, height = img.size 
    textwidth, textheight = d.textsize(text, font)
    x=width-textwidth-1
    y=height-textheight-1

    #Applying Text
    d.text((x,y), text, fill=(255,255,255, 200), font=font)

    #Combining Original Image with Text and Saving
    watermarked = Image.alpha_composite(img, txt)
    watermarked.save(output_file)

import sys
sys.path.append("clipit")

#@title Define generate function
import torch
import clipit
from datetime import datetime

import firebase_admin
from firebase_admin import credentials, firestore, storage

if not firebase_admin._apps:
    cred = credentials.Certificate("YOUR_CREDENTIAL")
    firebase_admin.initialize_app(cred, {
        'storageBucket': 'text2art.appspot.com'
    })

db = firestore.client()
bucket = storage.bucket()

def generate(doc_id, init_image, prompt, quality, style, aspect, email, public):
    torch.cuda.empty_cache()
    clipit.reset_settings()
    if init_image == '':
        init_image = None
    use_pixeldraw = (style == 'pixel art')
    use_clipdraw = (style == 'painting')
    seed = int(datetime.now().timestamp())
    enforce_smoothness = 50 if use_pixeldraw else 0
    enforce_saturation=1000 if use_pixeldraw else 0
    enforce_smoothness_type='log' if use_pixeldraw else 'default'
    clipit.add_settings(prompts=prompt,
                        init_image=init_image,
                        seed=seed,
                        aspect=aspect,
                        quality=quality,
                        scale=2.5,
                        display_every=7919,
                        use_pixeldraw=use_pixeldraw,
                        use_clipdraw=use_clipdraw,
                        enforce_smoothness=enforce_smoothness,
                        enforce_saturation=enforce_saturation,
                        enforce_smoothness_type=enforce_smoothness_type,
                        make_video=True)
    
    settings = clipit.apply_settings()
    clipit.do_init(settings)
    clipit.do_run(settings)
    watermark_image('output.png', 'output.png')


    image_blob = bucket.blob(f'generated_images/{doc_id}.png')
    image_blob.upload_from_filename('output.png')
    image_blob.make_public()

    video_blob = bucket.blob(f'generated_videos/{doc_id}.mp4')
    video_blob.upload_from_filename('output.mp4')
    video_blob.make_public()

    data = {
        "image": image_blob.public_url,
        "video": video_blob.public_url,
        "seed": seed,
        "prompt": prompt,
        "quality": quality,
        "aspect": aspect,
        "type": style,
        "user": email,
        "likes": 0,
        "public": public,
        "created_at": datetime.now()
    }
    db.collection('generated_images').document(doc_id).set(data)
    #email_results_sendgrid(email, prompt)
    email_results_mailgun(email, prompt)
    db.collection('emails').document(email).set({})
Imageio: 'ffmpeg-linux64-v3.3.1' was not found on your computer; downloading it now. Try 1. Download from https://github.com/imageio/imageio-binaries/raw/master/ffmpeg/ffmpeg-linux64-v3.3.1 (43.8 MB) Downloading: 8192/45929032 bytes (0.02220032/45929032 bytes (4.8%5267456/45929032 bytes (11.58413184/45929032 bytes (18.311444224/45929032 bytes (24.9%14630912/45929032 bytes (31.9%17801216/45929032 bytes (38.8%20914176/45929032 bytes (45.5%24133632/45929032 bytes (52.5%27418624/45929032 bytes (59.7%30482432/45929032 bytes (66.4%33677312/45929032 bytes (73.3%36675584/45929032 bytes (79.9%39952384/45929032 bytes (87.0%43098112/45929032 bytes (93.8%45929032/45929032 bytes (100.0%) Done File saved as /root/.imageio/ffmpeg/ffmpeg-linux64-v3.3.1. 
#@title Worker
import firebase_admin
from firebase_admin import credentials, storage, firestore
import time
from datetime import datetime

# Use the application default credentials
if not firebase_admin._apps:
    cred = credentials.Certificate("/content/text2art-firebase-adminsdk-xig66-1abd6a59a5.json")
    firebase_admin.initialize_app(cred, {
        'storageBucket': 'text2art.appspot.com'
    })

db = firestore.client()

db = firestore.client()
transaction = db.transaction()

@firestore.transactional
def claim_task(transaction, queue_objects_ref):
    # query firestore
    queue_objects = queue_objects_ref.stream(transaction=transaction)

    # pull the document from the iterable
    next_item = None
    for doc in queue_objects:
        next_item = doc

    # if queue is empty return status code of 2
    if not next_item:
        return {"status": 2}


    # get information from the document 
    next_item_data = next_item.to_dict()
    next_item_data["status"] = 0
    next_item_data['id'] = next_item.id

    # delete the document and return the information
    transaction.delete(next_item.reference)
    return next_item_data

# initialize query
queue_objects_ref = (
    db.collection("queue")
    .order_by("created_at", direction="ASCENDING")
    .limit(1)
)

transaction_attempts = 0
while True:
    try:
        # apply transaction
        next_item_data = claim_task(transaction, queue_objects_ref)
        if next_item_data['status'] == 0:
            generate(next_item_data['id'],
                     next_item_data['init_image'],
                     next_item_data['prompt'],
                     next_item_data['quality'],
                     next_item_data['type'],
                     next_item_data['aspect'],
                     next_item_data['email'],
                     next_item_data['public'])
            print(f"Generated {next_item_data['prompt']} for {next_item_data['email']}")

    except Exception as e:
        print(f"Could not apply transaction. Error: {e}")
        time.sleep(5)
        transaction_attempts += 1
        if transaction_attempts > 20:
            db.collection("errors").add({
                "exception": f"Could not apply artifaication claim transaction. Error: {e}",
                "location": "Claim artifaication",
                "time": str(datetime.now())
            })
            exit()
100%|███████████████████████████████████████| 338M/338M [00:06<00:00, 54.5MiB/s] 100%|███████████████████████████████████████| 335M/335M [00:05<00:00, 62.0MiB/s]
Using device: cuda:0 Optimising using: Adam Using text prompts: ['valorant'] Using seed: 1631366921
0it [00:00, ?it/s]
iter: 0, loss: 1.77756, losses: 0.896446, 0.881113, 0, -0
Image in a Jupyter notebook
iter: 10, loss: 1.72015, losses: 0.84635, 0.836922, 0.0409631, -0.00408081 iter: 20, loss: 1.69675, losses: 0.829035, 0.8129, 0.0626368, -0.00781748 iter: 30, loss: 1.64905, losses: 0.796871, 0.755306, 0.108576, -0.0117031 iter: 40, loss: 1.66617, losses: 0.793051, 0.763375, 0.124852, -0.015108 iter: 50, loss: 1.66109, losses: 0.785438, 0.748441, 0.145102, -0.0178915 iter: 60, loss: 1.61747, losses: 0.763514, 0.722728, 0.151594, -0.0203671 iter: 70, loss: 1.6243, losses: 0.756159, 0.717041, 0.173703, -0.0226 iter: 80, loss: 1.6517, losses: 0.755481, 0.721403, 0.20054, -0.0257214 iter: 90, loss: 1.63669, losses: 0.746971, 0.709676, 0.208813, -0.0287667 iter: 100, loss: 1.60978, losses: 0.710045, 0.662432, 0.26614, -0.0288405 iter: 110, loss: 1.64724, losses: 0.738359, 0.69756, 0.245333, -0.0340134 iter: 120, loss: 1.66847, losses: 0.74384, 0.705294, 0.254834, -0.0354999 iter: 130, loss: 1.67936, losses: 0.730546, 0.692289, 0.296178, -0.0396517 iter: 140, loss: 1.6723, losses: 0.732378, 0.690912, 0.293052, -0.0440403 iter: 150, loss: 1.7085, losses: 0.732171, 0.693373, 0.330178, -0.0472273 iter: 160, loss: 1.73284, losses: 0.741741, 0.705637, 0.337943, -0.0524856 iter: 170, loss: 1.6713, losses: 0.701596, 0.656513, 0.365261, -0.0520703 iter: 180, loss: 1.70569, losses: 0.696251, 0.652485, 0.415168, -0.0582116 iter: 190, loss: 1.70315, losses: 0.736267, 0.696805, 0.340532, -0.0704503 iter: 200, loss: 1.69851, losses: 0.727797, 0.686374, 0.362806, -0.0784678 iter: 210, loss: 1.76783, losses: 0.731699, 0.689574, 0.430919, -0.0843648 iter: 220, loss: 1.7265, losses: 0.704756, 0.660676, 0.444067, -0.083 iter: 230, loss: 1.78591, losses: 0.729457, 0.690979, 0.464429, -0.0989578 iter: 240, loss: 1.74899, losses: 0.703141, 0.66012, 0.482752, -0.0970195 iter: 250, loss: 1.75362, losses: 0.726844, 0.684422, 0.46179, -0.119441 iter: 260, loss: 1.77593, losses: 0.695332, 0.647767, 0.546829, -0.113994 iter: 270, loss: 1.77449, losses: 0.719728, 0.678615, 0.501977, -0.125828 iter: 280, loss: 1.81759, losses: 0.726844, 0.683753, 0.545979, -0.138987 iter: 290, loss: 1.82686, losses: 0.723786, 0.683685, 0.572579, -0.153194 iter: 300, loss: 1.77995, losses: 0.717711, 0.677354, 0.533758, -0.148869 iter: 310, loss: 1.79687, losses: 0.700435, 0.651312, 0.611562, -0.166442 iter: 320, loss: 1.8439, losses: 0.725378, 0.687535, 0.620812, -0.189828 iter: 330, loss: 1.84113, losses: 0.732268, 0.697751, 0.614871, -0.203756 iter: 340, loss: 1.80101, losses: 0.734152, 0.700013, 0.57725, -0.210407 iter: 350, loss: 1.87482, losses: 0.729735, 0.690156, 0.677581, -0.222648 [MoviePy] >>>> Building video output.mp4 [MoviePy] Writing video output.mp4
100%|██████████| 350/350 [00:03<00:00, 97.69it/s]
[MoviePy] Done. [MoviePy] >>>> Video ready: output.mp4 Generated valorant for [email protected] Using device: cuda:0 Optimising using: Adam Using text prompts: ['scenery of a classroom'] Using seed: 1631367441
0it [00:00, ?it/s]
iter: 0, loss: 1.79334, losses: 0.905605, 0.887731
Image in a Jupyter notebook
iter: 10, loss: 1.68093, losses: 0.844151, 0.836781 iter: 20, loss: 1.6049, losses: 0.799293, 0.805606 iter: 30, loss: 1.51132, losses: 0.75481, 0.756509 iter: 40, loss: 1.5133, losses: 0.757533, 0.755768 iter: 50, loss: 1.46864, losses: 0.738323, 0.730322 iter: 60, loss: 1.4448, losses: 0.725583, 0.719218 iter: 70, loss: 1.34917, losses: 0.682924, 0.66625 iter: 80, loss: 1.43634, losses: 0.724533, 0.711809 iter: 90, loss: 1.31944, losses: 0.66941, 0.650025 iter: 100, loss: 1.32222, losses: 0.669427, 0.652795 iter: 110, loss: 1.37596, losses: 0.697287, 0.678675 iter: 120, loss: 1.313, losses: 0.665578, 0.647419 iter: 130, loss: 1.30345, losses: 0.659978, 0.643476 iter: 140, loss: 1.2921, losses: 0.656035, 0.636068 iter: 150, loss: 1.37444, losses: 0.69543, 0.679006 iter: 160, loss: 1.29258, losses: 0.657204, 0.635373 iter: 170, loss: 1.36523, losses: 0.691643, 0.67359 iter: 180, loss: 1.36283, losses: 0.691069, 0.671757 iter: 190, loss: 1.34418, losses: 0.683817, 0.660367 iter: 200, loss: 1.34891, losses: 0.68595, 0.662964 iter: 210, loss: 1.34087, losses: 0.680077, 0.660789 iter: 220, loss: 1.28381, losses: 0.654696, 0.629117 iter: 230, loss: 1.34602, losses: 0.684541, 0.661475 iter: 240, loss: 1.38865, losses: 0.700425, 0.688222 iter: 250, loss: 1.35266, losses: 0.688038, 0.664627 iter: 260, loss: 1.28564, losses: 0.654066, 0.631576 iter: 270, loss: 1.35924, losses: 0.690176, 0.669061 iter: 280, loss: 1.349, losses: 0.685251, 0.663751 iter: 290, loss: 1.34651, losses: 0.684344, 0.662165 iter: 300, loss: 1.34307, losses: 0.684653, 0.658415 iter: 310, loss: 1.32193, losses: 0.674089, 0.647839 iter: 320, loss: 1.32965, losses: 0.678445, 0.651201 iter: 330, loss: 1.34759, losses: 0.685432, 0.662163 iter: 340, loss: 1.26004, losses: 0.644599, 0.615445 iter: 350, loss: 1.33461, losses: 0.680077, 0.65453 [MoviePy] >>>> Building video output.mp4 [MoviePy] Writing video output.mp4
100%|██████████| 350/350 [00:05<00:00, 63.88it/s]
[MoviePy] Done. [MoviePy] >>>> Video ready: output.mp4 Generated scenery of a classroom for [email protected] Using device: cuda:0 Optimising using: Adam Using text prompts: ['apes painting a mountain'] Using seed: 1631368244
0it [00:00, ?it/s]
iter: 0, loss: 1.83495, losses: 0.921623, 0.913329
Image in a Jupyter notebook
iter: 10, loss: 1.56493, losses: 0.800183, 0.764745 iter: 20, loss: 1.5014, losses: 0.769082, 0.732317 iter: 30, loss: 1.40154, losses: 0.719156, 0.682388 iter: 40, loss: 1.40957, losses: 0.725792, 0.683783 iter: 50, loss: 1.37559, losses: 0.713517, 0.662075 iter: 60, loss: 1.38377, losses: 0.719499, 0.66427 iter: 70, loss: 1.36002, losses: 0.709037, 0.65098 iter: 80, loss: 1.35096, losses: 0.706013, 0.644948 iter: 90, loss: 1.36741, losses: 0.715287, 0.65212 iter: 100, loss: 1.31304, losses: 0.68953, 0.623509 iter: 110, loss: 1.32172, losses: 0.699595, 0.622122 iter: 120, loss: 1.31483, losses: 0.694301, 0.620525 iter: 130, loss: 1.21858, losses: 0.660242, 0.558335 iter: 140, loss: 1.21807, losses: 0.65896, 0.559112 iter: 150, loss: 1.1966, losses: 0.649495, 0.547109 iter: 160, loss: 1.2718, losses: 0.679278, 0.592523 iter: 170, loss: 1.25779, losses: 0.669528, 0.588257 iter: 180, loss: 1.29333, losses: 0.688034, 0.605296 iter: 190, loss: 1.18015, losses: 0.648566, 0.531582 iter: 200, loss: 1.28123, losses: 0.683646, 0.597581 iter: 210, loss: 1.25582, losses: 0.673573, 0.582242 iter: 220, loss: 1.24113, losses: 0.669706, 0.571426 iter: 230, loss: 1.15053, losses: 0.636394, 0.514138 iter: 240, loss: 1.17241, losses: 0.648518, 0.523894 iter: 250, loss: 1.26389, losses: 0.679513, 0.584372 iter: 260, loss: 1.28481, losses: 0.690416, 0.594392 iter: 270, loss: 1.25473, losses: 0.673579, 0.581151 iter: 280, loss: 1.15301, losses: 0.635614, 0.517392 iter: 290, loss: 1.23887, losses: 0.670385, 0.568486 iter: 300, loss: 1.24428, losses: 0.668892, 0.575392 iter: 310, loss: 1.22025, losses: 0.657186, 0.563065 iter: 320, loss: 1.24961, losses: 0.67332, 0.576293 iter: 330, loss: 1.145, losses: 0.628816, 0.516186 iter: 340, loss: 1.23715, losses: 0.664273, 0.572875 iter: 350, loss: 1.22119, losses: 0.658177, 0.563016 [MoviePy] >>>> Building video output.mp4 [MoviePy] Writing video output.mp4
100%|██████████| 350/350 [00:07<00:00, 49.97it/s]
[MoviePy] Done. [MoviePy] >>>> Video ready: output.mp4 Generated apes painting a mountain for [email protected] Working with z of shape (1, 256, 16, 16) = 65536 dimensions.
Downloading: "https://download.pytorch.org/models/vgg16-397923af.pth" to /root/.cache/torch/hub/checkpoints/vgg16-397923af.pth
0%| | 0.00/528M [00:00<?, ?B/s]
loaded pretrained LPIPS loss from taming/modules/autoencoder/lpips/vgg.pth VQLPIPSWithDiscriminator running with hinge loss. Restored from models/vqgan_imagenet_f16_16384.ckpt Using device: cuda:0 Optimising using: Adam Using text prompts: ['nuclear explosion Artstation'] Using seed: 1631369036
0it [00:00, ?it/s]
iter: 0, loss: 1.98218, losses: 1.01732, 0.964859
Image in a Jupyter notebook
iter: 10, loss: 1.80967, losses: 0.934088, 0.875584 iter: 20, loss: 1.68411, losses: 0.866507, 0.817607 iter: 30, loss: 1.54657, losses: 0.787691, 0.75888 iter: 40, loss: 1.53726, losses: 0.791438, 0.74582 iter: 50, loss: 1.55067, losses: 0.798633, 0.752035 iter: 60, loss: 1.53845, losses: 0.791869, 0.746585 iter: 70, loss: 1.50005, losses: 0.766011, 0.734036 iter: 80, loss: 1.5107, losses: 0.773652, 0.737044 iter: 90, loss: 1.4061, losses: 0.718441, 0.687654 iter: 100, loss: 1.41059, losses: 0.720596, 0.689997 iter: 110, loss: 1.40362, losses: 0.715753, 0.687865 iter: 120, loss: 1.48753, losses: 0.762674, 0.724853 iter: 130, loss: 1.45812, losses: 0.748371, 0.709751 iter: 140, loss: 1.43548, losses: 0.734247, 0.70123 iter: 150, loss: 1.45012, losses: 0.73959, 0.71053 iter: 160, loss: 1.44826, losses: 0.738534, 0.709729 iter: 170, loss: 1.37094, losses: 0.693602, 0.677333 iter: 180, loss: 1.38504, losses: 0.702857, 0.682185 iter: 190, loss: 1.45619, losses: 0.74202, 0.71417 iter: 200, loss: 1.36881, losses: 0.699149, 0.669661 iter: 210, loss: 1.43083, losses: 0.73075, 0.700078 iter: 220, loss: 1.42584, losses: 0.730239, 0.695599 iter: 230, loss: 1.43937, losses: 0.732931, 0.706435 iter: 240, loss: 1.40205, losses: 0.710553, 0.691502 iter: 250, loss: 1.44059, losses: 0.729404, 0.711183 iter: 260, loss: 1.42367, losses: 0.717779, 0.705892 iter: 270, loss: 1.39675, losses: 0.70848, 0.688274 iter: 280, loss: 1.35778, losses: 0.684692, 0.673092 iter: 290, loss: 1.42198, losses: 0.726587, 0.69539 iter: 300, loss: 1.40248, losses: 0.711903, 0.690577 iter: 310, loss: 1.39789, losses: 0.70836, 0.689531 iter: 320, loss: 1.35195, losses: 0.686682, 0.665267 iter: 330, loss: 1.40928, losses: 0.72121, 0.688068 iter: 340, loss: 1.41663, losses: 0.718625, 0.698006 iter: 350, loss: 1.41958, losses: 0.72011, 0.699466 [MoviePy] >>>> Building video output.mp4 [MoviePy] Writing video output.mp4
100%|██████████| 350/350 [00:05<00:00, 69.15it/s]
[MoviePy] Done. [MoviePy] >>>> Video ready: output.mp4 Generated nuclear explosion Artstation for [email protected] Using device: cuda:0 Optimising using: Adam Using text prompts: ['money counter strike'] Using seed: 1631369366
0it [00:00, ?it/s]
iter: 0, loss: 1.86777, losses: 0.938345, 0.929424
Image in a Jupyter notebook
iter: 10, loss: 1.68013, losses: 0.841294, 0.83884 iter: 20, loss: 1.603, losses: 0.810142, 0.792857 iter: 30, loss: 1.49131, losses: 0.75163, 0.739675 iter: 40, loss: 1.50406, losses: 0.752236, 0.751819 iter: 50, loss: 1.49349, losses: 0.751017, 0.74247 iter: 60, loss: 1.4929, losses: 0.748544, 0.744359 iter: 70, loss: 1.4659, losses: 0.733613, 0.732289 iter: 80, loss: 1.43947, losses: 0.724887, 0.714587 iter: 90, loss: 1.39916, losses: 0.703197, 0.695961 iter: 100, loss: 1.37969, losses: 0.695542, 0.684148 iter: 110, loss: 1.45146, losses: 0.729388, 0.722072 iter: 120, loss: 1.41209, losses: 0.713261, 0.698827 iter: 130, loss: 1.42972, losses: 0.720548, 0.709171 iter: 140, loss: 1.3643, losses: 0.688804, 0.675492 iter: 150, loss: 1.42707, losses: 0.720021, 0.707046 iter: 160, loss: 1.40857, losses: 0.711569, 0.696999 iter: 170, loss: 1.34469, losses: 0.677766, 0.666927 iter: 180, loss: 1.38629, losses: 0.700613, 0.685673 iter: 190, loss: 1.38379, losses: 0.701282, 0.682506 iter: 200, loss: 1.39235, losses: 0.7013, 0.69105 iter: 210, loss: 1.36827, losses: 0.690302, 0.677971 iter: 220, loss: 1.36076, losses: 0.688686, 0.672073 iter: 230, loss: 1.31864, losses: 0.668321, 0.650314 iter: 240, loss: 1.32421, losses: 0.67418, 0.650032 iter: 250, loss: 1.36405, losses: 0.690226, 0.673829 iter: 260, loss: 1.30726, losses: 0.663133, 0.64413 iter: 270, loss: 1.29947, losses: 0.659359, 0.640107 iter: 280, loss: 1.30438, losses: 0.662969, 0.641415 iter: 290, loss: 1.31243, losses: 0.668021, 0.644408 iter: 300, loss: 1.34697, losses: 0.685241, 0.661731 iter: 310, loss: 1.35061, losses: 0.685719, 0.664892 iter: 320, loss: 1.34546, losses: 0.684394, 0.661068 iter: 330, loss: 1.35187, losses: 0.686155, 0.66571 iter: 340, loss: 1.34018, losses: 0.681595, 0.658583 iter: 350, loss: 1.3396, losses: 0.680151, 0.659445 [MoviePy] >>>> Building video output.mp4 [MoviePy] Writing video output.mp4
100%|██████████| 350/350 [00:07<00:00, 47.81it/s]
[MoviePy] Done. [MoviePy] >>>> Video ready: output.mp4 Generated money counter strike for [email protected] Using device: cuda:0 Optimising using: Adam Using text prompts: ['MONEY COUNTER STRIKE'] Using seed: 1631370153
0it [00:00, ?it/s]
iter: 0, loss: 1.86938, losses: 0.940531, 0.928845
Image in a Jupyter notebook
iter: 10, loss: 1.76271, losses: 0.885855, 0.876857 iter: 20, loss: 1.64056, losses: 0.826425, 0.814136 iter: 30, loss: 1.56893, losses: 0.786239, 0.782695 iter: 40, loss: 1.46572, losses: 0.736142, 0.729574 iter: 50, loss: 1.4132, losses: 0.710099, 0.703098 iter: 60, loss: 1.37555, losses: 0.693904, 0.681642 iter: 70, loss: 1.45161, losses: 0.730529, 0.721082 iter: 80, loss: 1.40499, losses: 0.707026, 0.697967 iter: 90, loss: 1.39186, losses: 0.699512, 0.692345 iter: 100, loss: 1.3345, losses: 0.674355, 0.66015 iter: 110, loss: 1.33217, losses: 0.672372, 0.659802 iter: 120, loss: 1.39448, losses: 0.704473, 0.690009 iter: 130, loss: 1.38477, losses: 0.701706, 0.683068 iter: 140, loss: 1.31858, losses: 0.673467, 0.645111 iter: 150, loss: 1.37422, losses: 0.698508, 0.675713 iter: 160, loss: 1.38322, losses: 0.704365, 0.678851 iter: 170, loss: 1.37846, losses: 0.700068, 0.678392 iter: 180, loss: 1.36329, losses: 0.69813, 0.665161 iter: 190, loss: 1.2743, losses: 0.658496, 0.615809 iter: 200, loss: 1.27397, losses: 0.659155, 0.614816 iter: 210, loss: 1.25968, losses: 0.65497, 0.604709 iter: 220, loss: 1.34366, losses: 0.689129, 0.65453 iter: 230, loss: 1.32478, losses: 0.682772, 0.642004 iter: 240, loss: 1.24918, losses: 0.651157, 0.598022 iter: 250, loss: 1.24449, losses: 0.651944, 0.592544 iter: 260, loss: 1.37471, losses: 0.705333, 0.669381 iter: 270, loss: 1.323, losses: 0.685068, 0.637935 iter: 280, loss: 1.37166, losses: 0.70388, 0.667781 iter: 290, loss: 1.23077, losses: 0.640115, 0.590658 iter: 300, loss: 1.32408, losses: 0.684972, 0.639113 iter: 310, loss: 1.33226, losses: 0.688485, 0.643774 iter: 320, loss: 1.32498, losses: 0.683852, 0.641129 iter: 330, loss: 1.33216, losses: 0.68877, 0.64339 iter: 340, loss: 1.36607, losses: 0.70164, 0.664431 iter: 350, loss: 1.33812, losses: 0.690416, 0.647707 [MoviePy] >>>> Building video output.mp4 [MoviePy] Writing video output.mp4
100%|██████████| 350/350 [00:07<00:00, 48.67it/s]
[MoviePy] Done. [MoviePy] >>>> Video ready: output.mp4 Generated MONEY COUNTER STRIKE for [email protected] Using device: cuda:0 Optimising using: Adam Using text prompts: ['Naruto singing Pixelart'] Using seed: 1631370965
0it [00:00, ?it/s]
iter: 0, loss: 0.919428, losses: 0.919428, 0, -0
Image in a Jupyter notebook
iter: 10, loss: 0.883077, losses: 0.853969, 0.0332053, -0.00409738 iter: 20, loss: 0.844653, losses: 0.76889, 0.0846299, -0.00886689 iter: 30, loss: 0.853612, losses: 0.765205, 0.101008, -0.0126013 iter: 40, loss: 0.842155, losses: 0.700698, 0.158412, -0.0169555 iter: 50, loss: 0.832684, losses: 0.668856, 0.183797, -0.0199688 iter: 60, loss: 0.844957, losses: 0.694442, 0.172256, -0.0217404 iter: 70, loss: 0.836571, losses: 0.635727, 0.225357, -0.0245124 iter: 80, loss: 0.822838, losses: 0.652828, 0.195229, -0.0252191 iter: 90, loss: 0.843595, losses: 0.652551, 0.218892, -0.0278486 iter: 100, loss: 0.839995, losses: 0.631504, 0.241655, -0.0331631 iter: 110, loss: 0.823319, losses: 0.587553, 0.270503, -0.0347358 iter: 120, loss: 0.85022, losses: 0.657152, 0.232321, -0.0392529 iter: 130, loss: 0.848634, losses: 0.588458, 0.305761, -0.0455852 iter: 140, loss: 0.906936, losses: 0.641672, 0.31988, -0.0546163 iter: 150, loss: 0.856862, losses: 0.630063, 0.287536, -0.0607366 iter: 160, loss: 0.89044, losses: 0.579484, 0.378261, -0.0673052 iter: 170, loss: 0.913349, losses: 0.644649, 0.346206, -0.0775054 iter: 180, loss: 0.881212, losses: 0.632485, 0.336189, -0.0874626 iter: 190, loss: 0.923592, losses: 0.57665, 0.435487, -0.0885445 iter: 200, loss: 0.923671, losses: 0.577522, 0.441644, -0.0954944 [MoviePy] >>>> Building video output.mp4 [MoviePy] Writing video output.mp4
100%|██████████| 200/200 [00:02<00:00, 95.90it/s]
[MoviePy] Done. [MoviePy] >>>> Video ready: output.mp4 Generated Naruto singing Pixelart for [email protected] Working with z of shape (1, 256, 16, 16) = 65536 dimensions. loaded pretrained LPIPS loss from taming/modules/autoencoder/lpips/vgg.pth VQLPIPSWithDiscriminator running with hinge loss. Restored from models/vqgan_imagenet_f16_16384.ckpt Using device: cuda:0 Optimising using: Adam Using text prompts: ['phoenix Painting Artstation'] Using seed: 1631371180
0it [00:00, ?it/s]
iter: 0, loss: 1.94513, losses: 0.981188, 0.963942
Image in a Jupyter notebook
iter: 10, loss: 1.69751, losses: 0.867004, 0.83051 iter: 20, loss: 1.57817, losses: 0.802645, 0.775526 iter: 30, loss: 1.44843, losses: 0.738531, 0.709895 iter: 40, loss: 1.48036, losses: 0.764014, 0.716342 iter: 50, loss: 1.41204, losses: 0.733407, 0.678628 iter: 60, loss: 1.44617, losses: 0.750603, 0.695572 iter: 70, loss: 1.40031, losses: 0.730187, 0.670126 iter: 80, loss: 1.42611, losses: 0.743005, 0.683109 iter: 90, loss: 1.33999, losses: 0.702955, 0.637033 iter: 100, loss: 1.41194, losses: 0.735777, 0.676165 iter: 110, loss: 1.33994, losses: 0.704838, 0.635098 iter: 120, loss: 1.38641, losses: 0.72717, 0.659239 iter: 130, loss: 1.3177, losses: 0.691892, 0.625813 iter: 140, loss: 1.36825, losses: 0.716562, 0.651685 iter: 150, loss: 1.347, losses: 0.70588, 0.641124 iter: 160, loss: 1.31678, losses: 0.693407, 0.623376 iter: 170, loss: 1.34033, losses: 0.704502, 0.635827 iter: 180, loss: 1.29917, losses: 0.682327, 0.616839 iter: 190, loss: 1.33643, losses: 0.702504, 0.633922 iter: 200, loss: 1.34989, losses: 0.707663, 0.642229 iter: 210, loss: 1.297, losses: 0.68455, 0.612445 iter: 220, loss: 1.35547, losses: 0.713422, 0.64205 iter: 230, loss: 1.35353, losses: 0.708495, 0.64503 iter: 240, loss: 1.33539, losses: 0.705139, 0.630253 iter: 250, loss: 1.28831, losses: 0.681418, 0.606892 iter: 260, loss: 1.36959, losses: 0.720375, 0.649217 iter: 270, loss: 1.33701, losses: 0.702668, 0.634341 iter: 280, loss: 1.33099, losses: 0.70268, 0.628314 iter: 290, loss: 1.27401, losses: 0.671226, 0.602782 iter: 300, loss: 1.36017, losses: 0.717412, 0.642757 iter: 310, loss: 1.27738, losses: 0.673247, 0.604132 iter: 320, loss: 1.33234, losses: 0.70406, 0.628277 iter: 330, loss: 1.33071, losses: 0.70125, 0.629463 iter: 340, loss: 1.33268, losses: 0.699602, 0.633076 iter: 350, loss: 1.32592, losses: 0.694325, 0.631592 [MoviePy] >>>> Building video output.mp4 [MoviePy] Writing video output.mp4
100%|██████████| 350/350 [00:07<00:00, 46.52it/s]
[MoviePy] Done. [MoviePy] >>>> Video ready: output.mp4 Generated phoenix Painting Artstation for [email protected] Working with z of shape (1, 256, 16, 16) = 65536 dimensions. loaded pretrained LPIPS loss from taming/modules/autoencoder/lpips/vgg.pth VQLPIPSWithDiscriminator running with hinge loss. Restored from models/vqgan_imagenet_f16_16384.ckpt Using device: cuda:0 Optimising using: Adam Using text prompts: ['Gold mountain'] Using seed: 1631371528
0it [00:00, ?it/s]
iter: 0, loss: 0.902389, losses: 0.902389
Image in a Jupyter notebook
iter: 10, loss: 0.824092, losses: 0.824092 iter: 20, loss: 0.809775, losses: 0.809775 iter: 30, loss: 0.791974, losses: 0.791974 iter: 40, loss: 0.786786, losses: 0.786786 iter: 50, loss: 0.778224, losses: 0.778224 iter: 60, loss: 0.748747, losses: 0.748747 iter: 70, loss: 0.747181, losses: 0.747181 iter: 80, loss: 0.763031, losses: 0.763031 iter: 90, loss: 0.766155, losses: 0.766155 iter: 100, loss: 0.762155, losses: 0.762155 iter: 110, loss: 0.748769, losses: 0.748769 iter: 120, loss: 0.749512, losses: 0.749512 iter: 130, loss: 0.722627, losses: 0.722627 iter: 140, loss: 0.717454, losses: 0.717454 iter: 150, loss: 0.714411, losses: 0.714411 iter: 160, loss: 0.740436, losses: 0.740436 iter: 170, loss: 0.706492, losses: 0.706492 iter: 180, loss: 0.739244, losses: 0.739244 iter: 190, loss: 0.745028, losses: 0.745028 iter: 200, loss: 0.726241, losses: 0.726241 [MoviePy] >>>> Building video output.mp4 [MoviePy] Writing video output.mp4
100%|██████████| 200/200 [00:03<00:00, 61.88it/s]
[MoviePy] Done. [MoviePy] >>>> Video ready: output.mp4 Generated Gold mountain for [email protected] Working with z of shape (1, 256, 16, 16) = 65536 dimensions. loaded pretrained LPIPS loss from taming/modules/autoencoder/lpips/vgg.pth VQLPIPSWithDiscriminator running with hinge loss. Restored from models/vqgan_imagenet_f16_16384.ckpt Using device: cuda:0 Optimising using: Adam Using text prompts: ['Asthetic'] Using seed: 1631371632
0it [00:00, ?it/s]
iter: 0, loss: 0.864731, losses: 0.864731
Image in a Jupyter notebook
iter: 10, loss: 0.822628, losses: 0.822628 iter: 20, loss: 0.815421, losses: 0.815421 iter: 30, loss: 0.802277, losses: 0.802277 iter: 40, loss: 0.802139, losses: 0.802139 iter: 50, loss: 0.800776, losses: 0.800776 iter: 60, loss: 0.792289, losses: 0.792289 iter: 70, loss: 0.794025, losses: 0.794025 iter: 80, loss: 0.793929, losses: 0.793929 iter: 90, loss: 0.788579, losses: 0.788579 iter: 100, loss: 0.788586, losses: 0.788586 iter: 110, loss: 0.784334, losses: 0.784334 iter: 120, loss: 0.780494, losses: 0.780494 iter: 130, loss: 0.782266, losses: 0.782266 iter: 140, loss: 0.782112, losses: 0.782112 iter: 150, loss: 0.769166, losses: 0.769166 iter: 160, loss: 0.766789, losses: 0.766789 iter: 170, loss: 0.780572, losses: 0.780572 iter: 180, loss: 0.763587, losses: 0.763587 iter: 190, loss: 0.785505, losses: 0.785505 iter: 200, loss: 0.777333, losses: 0.777333 [MoviePy] >>>> Building video output.mp4 [MoviePy] Writing video output.mp4
100%|██████████| 200/200 [00:03<00:00, 61.99it/s]
[MoviePy] Done. [MoviePy] >>>> Video ready: output.mp4 Generated Asthetic for [email protected] Working with z of shape (1, 256, 16, 16) = 65536 dimensions. loaded pretrained LPIPS loss from taming/modules/autoencoder/lpips/vgg.pth VQLPIPSWithDiscriminator running with hinge loss. Restored from models/vqgan_imagenet_f16_16384.ckpt Using device: cuda:0 Optimising using: Adam Using text prompts: ['Painting of bamboo forest #artstation'] Using seed: 1631372345
0it [00:00, ?it/s]
iter: 0, loss: 1.904, losses: 0.944618, 0.959383
Image in a Jupyter notebook
iter: 10, loss: 1.64256, losses: 0.813343, 0.829217 iter: 20, loss: 1.51725, losses: 0.748687, 0.768559 iter: 30, loss: 1.48822, losses: 0.744057, 0.744162 iter: 40, loss: 1.45277, losses: 0.735423, 0.717347 iter: 50, loss: 1.44284, losses: 0.725784, 0.717056 iter: 60, loss: 1.44916, losses: 0.725927, 0.723235 iter: 70, loss: 1.40935, losses: 0.710674, 0.698677 iter: 80, loss: 1.40663, losses: 0.707651, 0.69898 iter: 90, loss: 1.44602, losses: 0.725949, 0.720069 iter: 100, loss: 1.40984, losses: 0.705154, 0.704687 iter: 110, loss: 1.40817, losses: 0.702647, 0.705523 iter: 120, loss: 1.39691, losses: 0.700324, 0.696588 iter: 130, loss: 1.37356, losses: 0.687078, 0.686478 iter: 140, loss: 1.39843, losses: 0.696298, 0.702129 iter: 150, loss: 1.36154, losses: 0.687838, 0.673702 iter: 160, loss: 1.35182, losses: 0.678984, 0.672838 iter: 170, loss: 1.38376, losses: 0.693999, 0.689764 iter: 180, loss: 1.36542, losses: 0.684748, 0.680672 iter: 190, loss: 1.40212, losses: 0.706242, 0.695878 iter: 200, loss: 1.36864, losses: 0.690938, 0.677697 iter: 210, loss: 1.40633, losses: 0.70828, 0.698054 iter: 220, loss: 1.39427, losses: 0.699723, 0.694546 iter: 230, loss: 1.38626, losses: 0.699004, 0.687252 iter: 240, loss: 1.39298, losses: 0.700432, 0.692546 iter: 250, loss: 1.37379, losses: 0.690012, 0.683783 iter: 260, loss: 1.3707, losses: 0.68976, 0.680937 iter: 270, loss: 1.36751, losses: 0.690586, 0.676922 iter: 280, loss: 1.37026, losses: 0.687259, 0.683003 iter: 290, loss: 1.38236, losses: 0.693357, 0.688998 iter: 300, loss: 1.36359, losses: 0.685161, 0.67843 iter: 310, loss: 1.35293, losses: 0.680093, 0.672841 iter: 320, loss: 1.3723, losses: 0.692093, 0.680202 iter: 330, loss: 1.33728, losses: 0.672311, 0.664972 iter: 340, loss: 1.38506, losses: 0.695212, 0.689853 iter: 350, loss: 1.38134, losses: 0.695043, 0.686299 [MoviePy] >>>> Building video output.mp4 [MoviePy] Writing video output.mp4
100%|██████████| 350/350 [00:05<00:00, 59.54it/s]
[MoviePy] Done. [MoviePy] >>>> Video ready: output.mp4 Generated Painting of bamboo forest #artstation for [email protected] Working with z of shape (1, 256, 16, 16) = 65536 dimensions. loaded pretrained LPIPS loss from taming/modules/autoencoder/lpips/vgg.pth VQLPIPSWithDiscriminator running with hinge loss. Restored from models/vqgan_imagenet_f16_16384.ckpt Using device: cuda:0 Optimising using: Adam Using text prompts: ['Painting of earth burning Trending Artstation'] Using seed: 1631372911
0it [00:00, ?it/s]
iter: 0, loss: 1.9167, losses: 0.978897, 0.9378
Image in a Jupyter notebook
iter: 10, loss: 1.71482, losses: 0.885055, 0.829769 iter: 20, loss: 1.65738, losses: 0.860984, 0.796398 iter: 30, loss: 1.58763, losses: 0.829052, 0.758581 iter: 40, loss: 1.58654, losses: 0.826551, 0.759985 iter: 50, loss: 1.52198, losses: 0.788205, 0.73377 iter: 60, loss: 1.55257, losses: 0.810979, 0.74159 iter: 70, loss: 1.54849, losses: 0.807156, 0.741337 iter: 80, loss: 1.52862, losses: 0.79969, 0.728931 iter: 90, loss: 1.48795, losses: 0.770788, 0.717159 iter: 100, loss: 1.47139, losses: 0.758426, 0.712963 iter: 110, loss: 1.46631, losses: 0.756842, 0.709465 iter: 120, loss: 1.47045, losses: 0.763967, 0.706487 iter: 130, loss: 1.41774, losses: 0.728198, 0.689544 iter: 140, loss: 1.46891, losses: 0.755369, 0.713536 iter: 150, loss: 1.48085, losses: 0.768583, 0.712268 iter: 160, loss: 1.45829, losses: 0.754715, 0.703579 iter: 170, loss: 1.43316, losses: 0.737432, 0.695729 iter: 180, loss: 1.44927, losses: 0.748321, 0.700953 iter: 190, loss: 1.46645, losses: 0.759012, 0.707441 iter: 200, loss: 1.45402, losses: 0.751945, 0.702075 iter: 210, loss: 1.49582, losses: 0.773519, 0.722305 iter: 220, loss: 1.45426, losses: 0.750426, 0.703832 iter: 230, loss: 1.41239, losses: 0.72553, 0.686855 iter: 240, loss: 1.40029, losses: 0.718156, 0.682133 iter: 250, loss: 1.46503, losses: 0.763417, 0.701614 iter: 260, loss: 1.4054, losses: 0.727326, 0.678076 iter: 270, loss: 1.4444, losses: 0.747386, 0.697019 iter: 280, loss: 1.46755, losses: 0.757279, 0.710272 iter: 290, loss: 1.43157, losses: 0.738127, 0.693441 iter: 300, loss: 1.40702, losses: 0.725347, 0.681674 iter: 310, loss: 1.4764, losses: 0.759382, 0.717016 iter: 320, loss: 1.4664, losses: 0.764543, 0.701861 iter: 330, loss: 1.41394, losses: 0.734409, 0.679536 iter: 340, loss: 1.44464, losses: 0.743335, 0.701307 iter: 350, loss: 1.416, losses: 0.733011, 0.682994 [MoviePy] >>>> Building video output.mp4 [MoviePy] Writing video output.mp4
100%|██████████| 350/350 [00:05<00:00, 58.91it/s]
[MoviePy] Done. [MoviePy] >>>> Video ready: output.mp4 Generated Painting of earth burning Trending Artstation for [email protected] Working with z of shape (1, 256, 16, 16) = 65536 dimensions. loaded pretrained LPIPS loss from taming/modules/autoencoder/lpips/vgg.pth VQLPIPSWithDiscriminator running with hinge loss. Restored from models/vqgan_imagenet_f16_16384.ckpt Using device: cuda:0 Optimising using: Adam Using text prompts: ['Painting of falling meteor in space Artstation'] Using seed: 1631373401
0it [00:00, ?it/s]
iter: 0, loss: 2.02426, losses: 1.02631, 0.997955
Image in a Jupyter notebook
iter: 10, loss: 1.84575, losses: 0.936894, 0.908859 iter: 20, loss: 1.67538, losses: 0.851154, 0.824223 iter: 30, loss: 1.57974, losses: 0.802218, 0.777521 iter: 40, loss: 1.556, losses: 0.785031, 0.770971 iter: 50, loss: 1.48915, losses: 0.746871, 0.742279 iter: 60, loss: 1.55279, losses: 0.789324, 0.76347 iter: 70, loss: 1.53648, losses: 0.781755, 0.754725 iter: 80, loss: 1.46601, losses: 0.744377, 0.721636 iter: 90, loss: 1.51999, losses: 0.76914, 0.750847 iter: 100, loss: 1.46966, losses: 0.747058, 0.722604 iter: 110, loss: 1.4878, losses: 0.757247, 0.73055 iter: 120, loss: 1.46811, losses: 0.74646, 0.721652 iter: 130, loss: 1.47422, losses: 0.745233, 0.728985 iter: 140, loss: 1.47053, losses: 0.752861, 0.717671 iter: 150, loss: 1.47774, losses: 0.747446, 0.730293 iter: 160, loss: 1.46049, losses: 0.739541, 0.720949 iter: 170, loss: 1.45783, losses: 0.736865, 0.72097 iter: 180, loss: 1.45848, losses: 0.741493, 0.716987 iter: 190, loss: 1.40869, losses: 0.716256, 0.692434 iter: 200, loss: 1.46108, losses: 0.739957, 0.72112 iter: 210, loss: 1.44104, losses: 0.73062, 0.710423 iter: 220, loss: 1.47672, losses: 0.747214, 0.729505 iter: 230, loss: 1.40497, losses: 0.712324, 0.692642 iter: 240, loss: 1.40487, losses: 0.714787, 0.690083 iter: 250, loss: 1.43777, losses: 0.730997, 0.706769 iter: 260, loss: 1.44742, losses: 0.734575, 0.712842 iter: 270, loss: 1.39913, losses: 0.710018, 0.689117 iter: 280, loss: 1.41953, losses: 0.721556, 0.697978 iter: 290, loss: 1.39854, losses: 0.710491, 0.688054 iter: 300, loss: 1.40558, losses: 0.710153, 0.695429 iter: 310, loss: 1.45804, losses: 0.737635, 0.720403 iter: 320, loss: 1.46828, losses: 0.744884, 0.723395 iter: 330, loss: 1.43459, losses: 0.726583, 0.708004 iter: 340, loss: 1.4677, losses: 0.743237, 0.72446 iter: 350, loss: 1.39603, losses: 0.710689, 0.685342 [MoviePy] >>>> Building video output.mp4 [MoviePy] Writing video output.mp4
100%|██████████| 350/350 [00:05<00:00, 64.73it/s]
[MoviePy] Done. [MoviePy] >>>> Video ready: output.mp4 Generated Painting of falling meteor in space Artstation for [email protected] Using device: cuda:0 Optimising using: Adam Using text prompts: ['Global warming Artstation'] Using seed: 1631373820
0it [00:00, ?it/s]
iter: 0, loss: 1.91739, losses: 0.947136, 0.970257
Image in a Jupyter notebook
iter: 10, loss: 1.79696, losses: 0.882043, 0.914914 iter: 20, loss: 1.70994, losses: 0.838976, 0.870964 iter: 30, loss: 1.65716, losses: 0.821424, 0.835741 iter: 40, loss: 1.61085, losses: 0.802633, 0.808219 iter: 50, loss: 1.57866, losses: 0.793486, 0.785172 iter: 60, loss: 1.55191, losses: 0.778625, 0.773287 iter: 70, loss: 1.55137, losses: 0.777862, 0.773508 iter: 80, loss: 1.51771, losses: 0.761921, 0.755791 iter: 90, loss: 1.51008, losses: 0.760445, 0.749631 iter: 100, loss: 1.50085, losses: 0.753508, 0.747344 iter: 110, loss: 1.4831, losses: 0.747208, 0.735895 iter: 120, loss: 1.42988, losses: 0.721974, 0.707911 iter: 130, loss: 1.41534, losses: 0.715241, 0.700103 iter: 140, loss: 1.50022, losses: 0.759301, 0.740916 iter: 150, loss: 1.47029, losses: 0.744323, 0.725964 iter: 160, loss: 1.43192, losses: 0.723649, 0.708266 iter: 170, loss: 1.42465, losses: 0.719802, 0.704849 iter: 180, loss: 1.47325, losses: 0.74419, 0.729061 iter: 190, loss: 1.41029, losses: 0.713854, 0.696435 iter: 200, loss: 1.46393, losses: 0.742137, 0.72179 iter: 210, loss: 1.46823, losses: 0.74386, 0.724375 iter: 220, loss: 1.45808, losses: 0.740415, 0.717669 iter: 230, loss: 1.45798, losses: 0.737977, 0.719999 iter: 240, loss: 1.4255, losses: 0.71985, 0.705651 iter: 250, loss: 1.45257, losses: 0.73505, 0.717523 iter: 260, loss: 1.46726, losses: 0.740592, 0.726665 iter: 270, loss: 1.39017, losses: 0.708988, 0.68118 iter: 280, loss: 1.47692, losses: 0.748569, 0.728355 iter: 290, loss: 1.38534, losses: 0.70479, 0.68055 iter: 300, loss: 1.45064, losses: 0.733791, 0.716848 iter: 310, loss: 1.4602, losses: 0.737105, 0.723098 iter: 320, loss: 1.41909, losses: 0.722812, 0.696279 iter: 330, loss: 1.44316, losses: 0.732007, 0.711149 iter: 340, loss: 1.43294, losses: 0.723541, 0.709394 iter: 350, loss: 1.37639, losses: 0.700595, 0.675798 [MoviePy] >>>> Building video output.mp4 [MoviePy] Writing video output.mp4
100%|██████████| 350/350 [00:05<00:00, 60.32it/s]
[MoviePy] Done. [MoviePy] >>>> Video ready: output.mp4 Generated Global warming Artstation for [email protected] Working with z of shape (1, 256, 16, 16) = 65536 dimensions. loaded pretrained LPIPS loss from taming/modules/autoencoder/lpips/vgg.pth VQLPIPSWithDiscriminator running with hinge loss. Restored from models/vqgan_imagenet_f16_16384.ckpt Using device: cuda:0 Optimising using: Adam Using text prompts: ['Painting of refugee in war Artstation'] Using seed: 1631374715
0it [00:00, ?it/s]
iter: 0, loss: 1.97628, losses: 0.998302, 0.977976
Image in a Jupyter notebook
iter: 10, loss: 1.77938, losses: 0.892845, 0.886536 iter: 20, loss: 1.65414, losses: 0.827577, 0.826564 iter: 30, loss: 1.60201, losses: 0.808347, 0.793667 iter: 40, loss: 1.59222, losses: 0.800541, 0.791676 iter: 50, loss: 1.52639, losses: 0.766322, 0.760063 iter: 60, loss: 1.52387, losses: 0.768806, 0.755063 iter: 70, loss: 1.5179, losses: 0.766444, 0.751454 iter: 80, loss: 1.48272, losses: 0.744741, 0.737975 iter: 90, loss: 1.47441, losses: 0.745757, 0.72865 iter: 100, loss: 1.47521, losses: 0.738347, 0.736867 iter: 110, loss: 1.47067, losses: 0.744329, 0.726337 iter: 120, loss: 1.42496, losses: 0.72225, 0.702706 iter: 130, loss: 1.45384, losses: 0.733153, 0.720688 iter: 140, loss: 1.42747, losses: 0.722535, 0.704933 iter: 150, loss: 1.39226, losses: 0.708176, 0.684086 iter: 160, loss: 1.43324, losses: 0.722475, 0.710764 iter: 170, loss: 1.44782, losses: 0.727648, 0.720167 iter: 180, loss: 1.384, losses: 0.70132, 0.682677 iter: 190, loss: 1.4325, losses: 0.726302, 0.706197 iter: 200, loss: 1.43334, losses: 0.727539, 0.705805 iter: 210, loss: 1.42566, losses: 0.712839, 0.712818 iter: 220, loss: 1.37929, losses: 0.702746, 0.676545 iter: 230, loss: 1.41482, losses: 0.71554, 0.69928 iter: 240, loss: 1.39604, losses: 0.708121, 0.687921 iter: 250, loss: 1.4186, losses: 0.716632, 0.701972 iter: 260, loss: 1.43787, losses: 0.726257, 0.711618 iter: 270, loss: 1.38413, losses: 0.705033, 0.679097 iter: 280, loss: 1.36803, losses: 0.690475, 0.67755 iter: 290, loss: 1.43254, losses: 0.723743, 0.708802 iter: 300, loss: 1.36847, losses: 0.692327, 0.676138 iter: 310, loss: 1.44411, losses: 0.728248, 0.715867 iter: 320, loss: 1.37474, losses: 0.69589, 0.67885 iter: 330, loss: 1.35894, losses: 0.68735, 0.671592 iter: 340, loss: 1.41417, losses: 0.717271, 0.696896 iter: 350, loss: 1.37827, losses: 0.696367, 0.681906 [MoviePy] >>>> Building video output.mp4 [MoviePy] Writing video output.mp4
100%|██████████| 350/350 [00:05<00:00, 58.50it/s]
[MoviePy] Done. [MoviePy] >>>> Video ready: output.mp4 Generated Painting of refugee in war Artstation for [email protected] Working with z of shape (1, 256, 16, 16) = 65536 dimensions. loaded pretrained LPIPS loss from taming/modules/autoencoder/lpips/vgg.pth VQLPIPSWithDiscriminator running with hinge loss. Restored from models/vqgan_imagenet_f16_16384.ckpt Using device: cuda:0 Optimising using: Adam Using text prompts: ['Sky full of falling stars Artstation'] Using seed: 1631375843
0it [00:00, ?it/s]
iter: 0, loss: 1.91701, losses: 0.967604, 0.949409
Image in a Jupyter notebook
iter: 10, loss: 1.78863, losses: 0.914005, 0.874624 iter: 20, loss: 1.72675, losses: 0.885833, 0.840914 iter: 30, loss: 1.67174, losses: 0.862245, 0.809494 iter: 40, loss: 1.57118, losses: 0.815225, 0.755959 iter: 50, loss: 1.55671, losses: 0.804405, 0.752302 iter: 60, loss: 1.57161, losses: 0.805275, 0.766332 iter: 70, loss: 1.56484, losses: 0.801754, 0.763082 iter: 80, loss: 1.53884, losses: 0.791059, 0.747781 iter: 90, loss: 1.49417, losses: 0.76094, 0.733225 iter: 100, loss: 1.49531, losses: 0.763981, 0.731327 iter: 110, loss: 1.50424, losses: 0.767909, 0.736335 iter: 120, loss: 1.51958, losses: 0.77698, 0.7426 iter: 130, loss: 1.50219, losses: 0.764142, 0.738052 iter: 140, loss: 1.49126, losses: 0.763564, 0.7277 iter: 150, loss: 1.50962, losses: 0.77198, 0.737644 iter: 160, loss: 1.46008, losses: 0.747562, 0.712517 iter: 170, loss: 1.45498, losses: 0.746871, 0.708108 iter: 180, loss: 1.54586, losses: 0.788851, 0.757009 iter: 190, loss: 1.50349, losses: 0.770935, 0.732559 iter: 200, loss: 1.44546, losses: 0.741816, 0.70364 iter: 210, loss: 1.51671, losses: 0.775504, 0.741203 iter: 220, loss: 1.47076, losses: 0.753961, 0.716798 iter: 230, loss: 1.49871, losses: 0.765426, 0.73328 iter: 240, loss: 1.49533, losses: 0.766922, 0.728408 iter: 250, loss: 1.4908, losses: 0.765516, 0.725286 iter: 260, loss: 1.43599, losses: 0.733209, 0.702777 iter: 270, loss: 1.4904, losses: 0.764058, 0.726343 iter: 280, loss: 1.49806, losses: 0.764825, 0.733237 iter: 290, loss: 1.47025, losses: 0.750122, 0.720129 iter: 300, loss: 1.49843, losses: 0.76738, 0.731054 iter: 310, loss: 1.44298, losses: 0.734628, 0.708353 iter: 320, loss: 1.47557, losses: 0.756329, 0.719237 iter: 330, loss: 1.49117, losses: 0.764714, 0.726457 iter: 340, loss: 1.51247, losses: 0.775286, 0.737183 iter: 350, loss: 1.44398, losses: 0.736334, 0.707646 [MoviePy] >>>> Building video output.mp4 [MoviePy] Writing video output.mp4
100%|██████████| 350/350 [00:05<00:00, 59.80it/s]
[MoviePy] Done. [MoviePy] >>>> Video ready: output.mp4 Generated Sky full of falling stars Artstation for [email protected]
100%|███████████████████████████████████████| 244M/244M [00:04<00:00, 62.5MiB/s]
Using device: cuda:0 Optimising using: Adam Using text prompts: ['Deep sea fishes #pixelart'] Using seed: 1631376432
0it [00:00, ?it/s]
iter: 0, loss: 2.80006, losses: 0.998, 0.899528, 0.902529, 0, -0
Image in a Jupyter notebook
iter: 10, loss: 2.64148, losses: 0.965144, 0.819978, 0.828024, 0.0321351, -0.00380483 iter: 20, loss: 2.51608, losses: 0.925726, 0.765675, 0.770959, 0.0613005, -0.00758498 iter: 30, loss: 2.36191, losses: 0.840563, 0.716011, 0.705389, 0.111174, -0.0112212 iter: 40, loss: 2.37758, losses: 0.850156, 0.709268, 0.708682, 0.123898, -0.0144265 iter: 50, loss: 2.3883, losses: 0.840908, 0.713606, 0.706721, 0.144949, -0.0178794 iter: 60, loss: 2.38233, losses: 0.830858, 0.700258, 0.696721, 0.175498, -0.0210029 iter: 70, loss: 2.30994, losses: 0.785556, 0.668055, 0.651345, 0.229323, -0.0243434 iter: 80, loss: 2.37457, losses: 0.817185, 0.687364, 0.675064, 0.222628, -0.0276681 iter: 90, loss: 2.3833, losses: 0.810865, 0.692539, 0.68026, 0.230653, -0.0310152 iter: 100, loss: 2.28913, losses: 0.754148, 0.6559, 0.638885, 0.274187, -0.0339857 iter: 110, loss: 2.36714, losses: 0.795413, 0.67829, 0.664978, 0.265881, -0.0374195 iter: 120, loss: 2.29083, losses: 0.745225, 0.647482, 0.623588, 0.315621, -0.0410817 iter: 130, loss: 2.40615, losses: 0.799705, 0.681646, 0.66998, 0.301081, -0.0462598 iter: 140, loss: 2.38202, losses: 0.783221, 0.663967, 0.649852, 0.33642, -0.0514413 iter: 150, loss: 2.32646, losses: 0.73899, 0.638158, 0.617307, 0.387196, -0.0551896 iter: 160, loss: 2.42041, losses: 0.791158, 0.679555, 0.659178, 0.349544, -0.0590234 iter: 170, loss: 2.37861, losses: 0.782417, 0.668198, 0.652039, 0.339641, -0.0636876 iter: 180, loss: 2.35198, losses: 0.736817, 0.637208, 0.614196, 0.43469, -0.0709262 iter: 190, loss: 2.38252, losses: 0.769596, 0.654072, 0.635158, 0.40174, -0.0780451 iter: 200, loss: 2.45158, losses: 0.774603, 0.651112, 0.633251, 0.474529, -0.081916 iter: 210, loss: 2.38912, losses: 0.741154, 0.637113, 0.613078, 0.48807, -0.0902932 iter: 220, loss: 2.43175, losses: 0.775931, 0.657413, 0.641822, 0.454156, -0.0975684 iter: 230, loss: 2.43885, losses: 0.773034, 0.652623, 0.634703, 0.481586, -0.103096 iter: 240, loss: 2.40124, losses: 0.733509, 0.630775, 0.608416, 0.538027, -0.109486 iter: 250, loss: 2.50591, losses: 0.781124, 0.657904, 0.644816, 0.54414, -0.122073 iter: 260, loss: 2.48545, losses: 0.779218, 0.661381, 0.644847, 0.525711, -0.125709 iter: 270, loss: 2.5538, losses: 0.788688, 0.664807, 0.650737, 0.583396, -0.133829 iter: 280, loss: 2.48245, losses: 0.773785, 0.66136, 0.647841, 0.544927, -0.145468 iter: 290, loss: 2.44289, losses: 0.731593, 0.637906, 0.61176, 0.61808, -0.156452 iter: 300, loss: 2.50609, losses: 0.738175, 0.637866, 0.61152, 0.682619, -0.16409 iter: 310, loss: 2.51202, losses: 0.766506, 0.654692, 0.640785, 0.619037, -0.168999 iter: 320, loss: 2.50454, losses: 0.774904, 0.666689, 0.643698, 0.586361, -0.167107 iter: 330, loss: 2.63025, losses: 0.802685, 0.678268, 0.66111, 0.670411, -0.182229 iter: 340, loss: 2.59727, losses: 0.741989, 0.642928, 0.616166, 0.792642, -0.196459 iter: 350, loss: 2.64004, losses: 0.775725, 0.662162, 0.642339, 0.759287, -0.199475 iter: 360, loss: 2.60076, losses: 0.788635, 0.672498, 0.649927, 0.693331, -0.203631 iter: 370, loss: 2.63265, losses: 0.778029, 0.662542, 0.64566, 0.759275, -0.212861 iter: 380, loss: 2.61271, losses: 0.743628, 0.636124, 0.611075, 0.852561, -0.230682 iter: 390, loss: 2.47165, losses: 0.730847, 0.632052, 0.606626, 0.732472, -0.230347 iter: 400, loss: 2.58381, losses: 0.741026, 0.637247, 0.612336, 0.836275, -0.243077 iter: 410, loss: 2.63533, losses: 0.739657, 0.638103, 0.615026, 0.900662, -0.258121 iter: 420, loss: 2.65279, losses: 0.745474, 0.64463, 0.614185, 0.910191, -0.261691 iter: 430, loss: 2.59776, losses: 0.740172, 0.637207, 0.613903, 0.869276, -0.262799 iter: 440, loss: 2.64867, losses: 0.739976, 0.632641, 0.609864, 0.943959, -0.277772 iter: 450, loss: 2.66461, losses: 0.743323, 0.635771, 0.608924, 0.950076, -0.273479 iter: 460, loss: 2.66024, losses: 0.73997, 0.638964, 0.613162, 0.961503, -0.293359 iter: 470, loss: 2.67023, losses: 0.742876, 0.640083, 0.61359, 0.968357, -0.294671 iter: 480, loss: 2.61025, losses: 0.777279, 0.664377, 0.643243, 0.813982, -0.288631 iter: 490, loss: 2.58599, losses: 0.773732, 0.659844, 0.638298, 0.811209, -0.297093 iter: 500, loss: 2.83057, losses: 0.791359, 0.668875, 0.651, 1.0378, -0.318466 [MoviePy] >>>> Building video output.mp4 [MoviePy] Writing video output.mp4
100%|██████████| 500/500 [00:06<00:00, 82.49it/s]
[MoviePy] Done. [MoviePy] >>>> Video ready: output.mp4 Generated Deep sea fishes #pixelart for [email protected] Using device: cuda:0 Optimising using: Adam Using text prompts: ['Desert temple #pixelart'] Using seed: 1631379468
0it [00:00, ?it/s]
iter: 0, loss: 2.86178, losses: 0.995432, 0.918206, 0.948139, 0, -0
Image in a Jupyter notebook
iter: 10, loss: 2.65664, losses: 0.948724, 0.831039, 0.840647, 0.0402595, -0.00402742 iter: 20, loss: 2.5791, losses: 0.913788, 0.796731, 0.814007, 0.0619277, -0.00735546 iter: 30, loss: 2.52579, losses: 0.879295, 0.772825, 0.785154, 0.099673, -0.0111617 iter: 40, loss: 2.45592, losses: 0.854862, 0.743347, 0.744708, 0.126577, -0.0135733 iter: 50, loss: 2.4211, losses: 0.837284, 0.723882, 0.727294, 0.148886, -0.0162431 iter: 60, loss: 2.39617, losses: 0.822666, 0.703227, 0.705974, 0.183653, -0.019351 iter: 70, loss: 2.42274, losses: 0.821785, 0.705605, 0.713454, 0.203923, -0.0220318 iter: 80, loss: 2.3926, losses: 0.804075, 0.686107, 0.691539, 0.235266, -0.0243921 iter: 90, loss: 2.31298, losses: 0.751533, 0.660057, 0.652989, 0.27496, -0.0265601 iter: 100, loss: 2.3891, losses: 0.782347, 0.681727, 0.686485, 0.266955, -0.0284143 iter: 110, loss: 2.37228, losses: 0.786015, 0.678607, 0.674538, 0.262973, -0.0298529 iter: 120, loss: 2.35841, losses: 0.764385, 0.668929, 0.662699, 0.295793, -0.0333917 iter: 130, loss: 2.30664, losses: 0.727471, 0.646515, 0.629908, 0.337838, -0.0350897 iter: 140, loss: 2.32061, losses: 0.730319, 0.649038, 0.629088, 0.349199, -0.037033 iter: 150, loss: 2.37274, losses: 0.75671, 0.670303, 0.656566, 0.329373, -0.0402065 iter: 160, loss: 2.4238, losses: 0.771775, 0.675335, 0.66312, 0.358729, -0.045164 iter: 170, loss: 2.43526, losses: 0.786168, 0.676308, 0.671416, 0.348842, -0.0474716 iter: 180, loss: 2.36553, losses: 0.71788, 0.64842, 0.618985, 0.430025, -0.0497826 iter: 190, loss: 2.44266, losses: 0.773778, 0.672178, 0.665113, 0.383926, -0.0523398 iter: 200, loss: 2.48309, losses: 0.762021, 0.665317, 0.64231, 0.472961, -0.0595179 iter: 210, loss: 2.43828, losses: 0.750946, 0.66337, 0.647063, 0.437513, -0.060612 iter: 220, loss: 2.46016, losses: 0.751308, 0.673437, 0.649599, 0.450229, -0.0644123 iter: 230, loss: 2.50448, losses: 0.775825, 0.680905, 0.667939, 0.446709, -0.0669008 iter: 240, loss: 2.44087, losses: 0.713932, 0.648351, 0.610088, 0.541359, -0.072857 iter: 250, loss: 2.4737, losses: 0.750186, 0.664187, 0.647218, 0.49058, -0.0784767 iter: 260, loss: 2.48562, losses: 0.719965, 0.654174, 0.615551, 0.576247, -0.0803175 iter: 270, loss: 2.59375, losses: 0.772891, 0.677657, 0.650873, 0.580942, -0.0886092 iter: 280, loss: 2.50677, losses: 0.727562, 0.654889, 0.61696, 0.596202, -0.0888465 iter: 290, loss: 2.56675, losses: 0.728305, 0.657262, 0.620853, 0.654908, -0.0945767 iter: 300, loss: 2.64923, losses: 0.779187, 0.681686, 0.661825, 0.630893, -0.104359 iter: 310, loss: 2.57889, losses: 0.755511, 0.665925, 0.648101, 0.61666, -0.107305 iter: 320, loss: 2.59994, losses: 0.758091, 0.67061, 0.64705, 0.637194, -0.11301 iter: 330, loss: 2.6486, losses: 0.762261, 0.670162, 0.647517, 0.685922, -0.117266 iter: 340, loss: 2.626, losses: 0.776903, 0.674826, 0.65872, 0.638801, -0.12325 iter: 350, loss: 2.61157, losses: 0.752595, 0.66711, 0.645391, 0.673898, -0.127424 iter: 360, loss: 2.70205, losses: 0.770346, 0.669578, 0.647979, 0.751936, -0.137785 iter: 370, loss: 2.61834, losses: 0.723386, 0.646287, 0.606273, 0.780097, -0.137706 iter: 380, loss: 2.60966, losses: 0.737401, 0.652972, 0.612425, 0.750028, -0.14317 iter: 390, loss: 2.6515, losses: 0.76649, 0.673637, 0.652367, 0.709221, -0.150211 iter: 400, loss: 2.67669, losses: 0.773918, 0.664533, 0.641009, 0.758726, -0.161496 iter: 410, loss: 2.71786, losses: 0.763147, 0.664804, 0.645693, 0.811805, -0.167589 iter: 420, loss: 2.60568, losses: 0.734961, 0.653419, 0.612123, 0.773353, -0.168177 iter: 430, loss: 2.58263, losses: 0.760616, 0.664407, 0.63835, 0.694995, -0.175736 iter: 440, loss: 2.78597, losses: 0.78205, 0.681134, 0.662743, 0.853007, -0.19296 iter: 450, loss: 2.79606, losses: 0.781684, 0.668796, 0.649213, 0.900326, -0.203961 iter: 460, loss: 2.70951, losses: 0.775402, 0.673625, 0.648677, 0.820758, -0.208952 iter: 470, loss: 2.83391, losses: 0.748961, 0.65512, 0.616707, 1.02082, -0.207699 iter: 480, loss: 2.80554, losses: 0.776886, 0.676186, 0.6565, 0.921164, -0.225198 iter: 490, loss: 2.84207, losses: 0.791634, 0.675909, 0.656064, 0.960325, -0.241865 iter: 500, loss: 2.73522, losses: 0.773705, 0.670682, 0.644633, 0.888733, -0.242532 [MoviePy] >>>> Building video output.mp4 [MoviePy] Writing video output.mp4
100%|██████████| 500/500 [00:06<00:00, 80.16it/s]
[MoviePy] Done. [MoviePy] >>>> Video ready: output.mp4 Generated Desert temple #pixelart for [email protected] Using device: cuda:0 Optimising using: Adam Using text prompts: ['Medieval kingdom in fantasy videogame #pixelart'] Using seed: 1631380365
0it [00:00, ?it/s]
iter: 0, loss: 2.86872, losses: 1.00333, 0.930705, 0.934681, 0, -0
Image in a Jupyter notebook
iter: 10, loss: 2.71407, losses: 0.953444, 0.86192, 0.869894, 0.0327056, -0.00389109 iter: 20, loss: 2.57569, losses: 0.904329, 0.801233, 0.810879, 0.0667163, -0.00746464 iter: 30, loss: 2.50871, losses: 0.87019, 0.779682, 0.781287, 0.0881834, -0.0106303 iter: 40, loss: 2.4828, losses: 0.857213, 0.758231, 0.765787, 0.115199, -0.0136312 iter: 50, loss: 2.43499, losses: 0.833987, 0.734452, 0.741435, 0.141674, -0.0165585 iter: 60, loss: 2.45022, losses: 0.830956, 0.730547, 0.733725, 0.17434, -0.0193461 iter: 70, loss: 2.42769, losses: 0.812391, 0.726648, 0.72787, 0.182197, -0.0214158 iter: 80, loss: 2.43402, losses: 0.814304, 0.72248, 0.725145, 0.195131, -0.0230382 iter: 90, loss: 2.41639, losses: 0.790623, 0.700048, 0.702347, 0.24951, -0.0261411 iter: 100, loss: 2.36618, losses: 0.76295, 0.685952, 0.678307, 0.266095, -0.0271284 iter: 110, loss: 2.39427, losses: 0.761523, 0.683658, 0.674659, 0.304659, -0.0302333 iter: 120, loss: 2.41363, losses: 0.753141, 0.683062, 0.672473, 0.338536, -0.0335797 iter: 130, loss: 2.47287, losses: 0.80162, 0.708559, 0.706239, 0.291306, -0.0348553 iter: 140, loss: 2.40494, losses: 0.74693, 0.672786, 0.66008, 0.364261, -0.03912 iter: 150, loss: 2.48999, losses: 0.78822, 0.693787, 0.694232, 0.355145, -0.0413913 iter: 160, loss: 2.53118, losses: 0.797213, 0.701178, 0.69974, 0.378159, -0.045108 iter: 170, loss: 2.44233, losses: 0.751039, 0.67879, 0.663963, 0.396218, -0.0476786 iter: 180, loss: 2.57229, losses: 0.792922, 0.706616, 0.702951, 0.421496, -0.0516966 iter: 190, loss: 2.54006, losses: 0.795504, 0.709629, 0.702626, 0.385784, -0.0534833 iter: 200, loss: 2.50304, losses: 0.766359, 0.687359, 0.678544, 0.428573, -0.0577918 iter: 210, loss: 2.56448, losses: 0.788303, 0.701177, 0.693692, 0.445393, -0.064088 iter: 220, loss: 2.58568, losses: 0.783134, 0.696916, 0.690822, 0.481293, -0.0664818 iter: 230, loss: 2.61276, losses: 0.780535, 0.698189, 0.691806, 0.516213, -0.0739833 iter: 240, loss: 2.56739, losses: 0.780269, 0.70345, 0.693907, 0.467307, -0.0775405 iter: 250, loss: 2.60142, losses: 0.780329, 0.700847, 0.687612, 0.518334, -0.0856972 iter: 260, loss: 2.63075, losses: 0.749102, 0.675547, 0.662572, 0.637228, -0.0936983 iter: 270, loss: 2.56611, losses: 0.764391, 0.696047, 0.682235, 0.515543, -0.0921086 iter: 280, loss: 2.67199, losses: 0.779943, 0.695055, 0.686342, 0.612384, -0.101733 iter: 290, loss: 2.64289, losses: 0.768423, 0.690891, 0.681236, 0.612109, -0.109767 iter: 300, loss: 2.68104, losses: 0.758085, 0.682458, 0.665712, 0.699222, -0.124441 iter: 310, loss: 2.68655, losses: 0.77888, 0.698821, 0.683688, 0.651457, -0.126298 iter: 320, loss: 2.6422, losses: 0.792228, 0.710866, 0.703143, 0.571194, -0.135231 iter: 330, loss: 2.74777, losses: 0.793536, 0.705191, 0.701974, 0.695599, -0.148535 iter: 340, loss: 2.71528, losses: 0.752636, 0.680273, 0.665033, 0.783052, -0.165711 iter: 350, loss: 2.79262, losses: 0.791423, 0.708945, 0.70331, 0.753387, -0.16444 iter: 360, loss: 2.67425, losses: 0.77543, 0.698495, 0.690471, 0.671591, -0.16174 iter: 370, loss: 2.73424, losses: 0.775951, 0.699994, 0.689725, 0.747402, -0.178827 iter: 380, loss: 2.82747, losses: 0.793716, 0.708768, 0.705205, 0.807634, -0.187852 iter: 390, loss: 2.73897, losses: 0.784859, 0.712357, 0.696363, 0.744022, -0.198633 iter: 400, loss: 2.73399, losses: 0.789101, 0.709174, 0.703909, 0.726017, -0.194212 iter: 410, loss: 2.77963, losses: 0.791468, 0.712604, 0.701386, 0.788592, -0.214415 iter: 420, loss: 2.75013, losses: 0.774322, 0.699734, 0.690343, 0.80277, -0.21704 iter: 430, loss: 2.83816, losses: 0.777128, 0.698773, 0.689737, 0.901922, -0.229395 iter: 440, loss: 2.80265, losses: 0.785832, 0.71051, 0.70012, 0.84936, -0.243177 iter: 450, loss: 2.86095, losses: 0.789209, 0.710602, 0.700974, 0.899882, -0.239721 iter: 460, loss: 2.90074, losses: 0.791142, 0.713668, 0.708039, 0.92441, -0.236521 iter: 470, loss: 2.76619, losses: 0.778078, 0.702542, 0.693831, 0.8427, -0.250964 iter: 480, loss: 2.83438, losses: 0.782531, 0.70791, 0.696239, 0.914293, -0.266598 iter: 490, loss: 2.76304, losses: 0.748879, 0.683798, 0.666239, 0.953618, -0.289497 iter: 500, loss: 2.99032, losses: 0.797453, 0.715456, 0.704575, 1.05606, -0.283222 [MoviePy] >>>> Building video output.mp4 [MoviePy] Writing video output.mp4
100%|██████████| 500/500 [00:06<00:00, 82.94it/s]
[MoviePy] Done. [MoviePy] >>>> Video ready: output.mp4 Generated Medieval kingdom in fantasy videogame #pixelart for [email protected]