Doron Adler Norod
Norod
/ patch_for_wsl2_pytorch191_no-xformers_needs-disable-unsafe-pickle.diff
Created
October 26, 2022 12:53
Patch AUTOMATIC1111/stable-diffusion-webui to work on wsl2 with torch 1.9.1+cu111 (no xformers, needs disable-unsafe-pickle, no mem monitoring)
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| diff --git a/.gitignore b/.gitignore | |
| index 70660c5..6b51c8e 100644 | |
| --- a/.gitignore | |
| +++ b/.gitignore | |
| @@ -30,3 +30,6 @@ notification.mp3 | |
| .vscode | |
| /extensions | |
| +cache | |
| +gfpgan/ |
Norod
/ buildimagedatasetwithblipcaptionsanduploadtohf.ipynb
Created
November 3, 2022 11:26
BuildImageDatasetWithBlipCaptionsAndUploadToHF.ipynb
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Norod
/ hebrew-poem-generator-flaxgptneoforcausallm-inference-tpu.ipynb
Last active
January 26, 2025 12:26
hebrew-poem-generator-flaxgptneoforcausallm-inference-tpu.ipynb
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Norod
/ DPMSolverMultistepScheduler-stable-cpu.py
Created
November 8, 2022 14:11
CPU stable-diffusion-v1-5 inference using DPMSolverMultistepScheduler
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| from diffusers import StableDiffusionPipeline, DPMSolverMultistepScheduler | |
| import torch | |
| def main(): | |
| #//////////////////////////////////////////// | |
| seed = 42 | |
| model = "runwayml/stable-diffusion-v1-5" | |
| #//////////////////////////////////////////// |
Norod
/ sd-simpsons-model-DPMSolverMultistepScheduler.py
Created
November 10, 2022 20:31
Inference for sd-simpsons-model using DPMSolverMultistepScheduler
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| from diffusers import StableDiffusionPipeline, DPMSolverMultistepScheduler | |
| import torch | |
| def main(): | |
| #//////////////////////////////////////////// | |
| seed = 42 | |
| model = "Norod78/sd-simpsons-model" | |
| #//////////////////////////////////////////// |
Norod
/ sd-simpsons-model-img2img.py
Last active
December 1, 2022 11:33
Like the default HF StableDiffusion img2img sample, only it uses the sd2-simpsons-blip and DPMSolverMultistepScheduler
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| import requests | |
| import torch | |
| from PIL import Image | |
| from io import BytesIO | |
| from diffusers import StableDiffusionImg2ImgPipeline, DPMSolverMultistepScheduler | |
| def main(): | |
| #//////////////////////////////////////////// | |
| seed = 42 |
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Norod
/ two_tigers-StableDiffusionDepth2ImgPipeline.py
Created
December 9, 2022 17:15
The Huggingface example code for stabilityai/stable-diffusion-2-depth (with GPU, CPU and multiple image results)
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| import torch | |
| import requests | |
| from PIL import Image | |
| from diffusers import StableDiffusionDepth2ImgPipeline | |
| def main(): | |
| device = "cuda" if torch.cuda.is_available() else "cpu" | |
| dtype = torch.float16 if device == "cuda" else torch.float32 | |
| pipe = StableDiffusionDepth2ImgPipeline.from_pretrained( |
Norod
/ heb_eng_tokenize_compare.py
Created
December 26, 2022 09:41
Compare the amount of tokens generated for a given Hebrew word input when an English and Hebrew tokenizers are used
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| from transformers import AutoTokenizer | |
| tokenizer_heb = AutoTokenizer.from_pretrained("Norod78/hebrew-gpt_neo-small") | |
| tokenizer_eng = AutoTokenizer.from_pretrained("gpt2") | |
| prompt_text="שלום" | |
| prompt_length = len(prompt_text) | |
| encoded_prompt_heb = tokenizer_heb.encode(prompt_text, add_special_tokens=False, return_tensors="pt") | |
| num_of_tokenz_heb = encoded_prompt_heb.size()[-1] | |
| print(f"Hebrew tokenizer: Tokens = {num_of_tokenz_heb} length = {prompt_length}") #Hebrew tokenizer: Tokens = 1 length = 4 | |
| encoded_prompt_eng = tokenizer_eng.encode(prompt_text, add_special_tokens=False, return_tensors="pt") |
Norod
/ sd-diffusers-model-compare.py
Last active
July 29, 2023 20:12
Compare two Stable-Diffusion diffuser models using a predifined set of prompts
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| from diffusers import StableDiffusionPipeline, DPMSolverMultistepScheduler | |
| import torch | |
| #//////////////////////////////////////////////////////////////// | |
| guidance_scale=8.0 | |
| steps=40 | |
| width=512 | |
| height=512 |