Sayak Paul sayakpaul

## benchmark_pixart-900m-1024-ft.py
import torch

torch.set_float32_matmul_precision("high")

from diffusers import DiffusionPipeline
import time

pipeline_id = "ptx0/pixart-900m-1024-ft"
pipeline = DiffusionPipeline.from_pretrained(
    pipeline_id,

## run_sd3_compile.py
import torch

torch.set_float32_matmul_precision("high")

from diffusers import StableDiffusion3Pipeline
import time

id = "stabilityai/stable-diffusion-3-medium-diffusers"
pipeline = StableDiffusion3Pipeline.from_pretrained(
    id,

## run_sd3_8bit.py
from diffusers import StableDiffusion3Pipeline
from transformers import T5EncoderModel
import torch
import time
import gc

def flush():
    gc.collect()
    torch.cuda.empty_cache()

## run_hunyuan_dit_compile.py
import torch

torch.set_float32_matmul_precision("high")

from diffusers import HunyuanDiTPipeline
import argparse
import time


def load_pipeline(args):

## run_hunyuan_dit_less_memory.py
"""
Make sure you have `diffusers`, `accelerate`, `transformers`, and `bitsandbytes` installed.

You also set up PyTorch and CUDA.

Once the dependencies are installed, you can run `python run_hunyuan_dit_less_memory.py`.
"""

from diffusers import HunyuanDiTPipeline
from transformers import T5EncoderModel

## custom_pipelines_with_custom_transformers.ipynb

      
              1 file
            
          
              0 forks
            
          
              0 comments
            
          
              0 stars
            
          
                sayakpaul
                / custom_pipelines_with_custom_transformers.ipynb
            
            
              Created
              March 13, 2024 12:03
            
              
                This notebook shows how to use a custom `transformers` model within a Stable Diffusion pipeline. 
              
          
        Loading

      Sorry, something went wrong. Reload?
      Sorry, we cannot display this file.
      Sorry, this file is invalid so it cannot be displayed.
      
          Viewer requires iframe.
      
    
## instructions.md

      
              1 file
            
          
              0 forks
            
          
              0 comments
            
          
              1 star
            
          
                sayakpaul
                / instructions.md
            
            
              Last active
              February 15, 2024 18:23
            
              
                Utilities for formatting Python projects. Inspired from what's done in `diffusers`. 
              
          
Get a Makefile from here: https://github.com/huggingface/diffusion-fast/blob/main/Makefile.
Install ruff: pip install ruff.
Get the ruff configuration file through pyproject.toml: https://github.com/huggingface/diffusion-fast/blob/main/pyproject.toml.

Keep both the Makefile and pyproject.toml files at the root level of your project.
Before staging your changes, run the following from the root of your project directory:
make style &amp;&amp; make quality

  
## coco_30k_hf_datasets.py
from datasets import Dataset, Features
from datasets import Image as ImageFeature
from datasets import Value
import pandas as pd
import os

# CSV comes from the notebook above.
df = pd.read_csv("coco_30k_randomly_sampled_2014_val.csv")
root_path = "val2014"

## get_params_diffusers_diffusers.py
from huggingface_hub import (
    hf_hub_download,
    list_files_info,
    parse_safetensors_file_metadata,
)
import json
import argparse


def get_files_info(args):

## check_logits_with_sd_peft_lora.py
# SDXL: 0.613, 0.5566, 0.54, 0.4162, 0.4042, 0.4596, 0.5374, 0.5286, 0.5038
# SD: 0.5396, 0.5707, 0.477, 0.4665, 0.5419, 0.4594, 0.4857, 0.4741, 0.4804

from diffusers import DiffusionPipeline
from peft import LoraConfig
import argparse
import torch


def load_pipeline(pipeline_id):
	import torch

	torch.set_float32_matmul_precision("high")

	from diffusers import DiffusionPipeline
	import time

	pipeline_id = "ptx0/pixart-900m-1024-ft"
	pipeline = DiffusionPipeline.from_pretrained(
	pipeline_id,
	import torch

	torch.set_float32_matmul_precision("high")

	from diffusers import StableDiffusion3Pipeline
	import time

	id = "stabilityai/stable-diffusion-3-medium-diffusers"
	pipeline = StableDiffusion3Pipeline.from_pretrained(
	id,
	from diffusers import StableDiffusion3Pipeline
	from transformers import T5EncoderModel
	import torch
	import time
	import gc

	def flush():
	gc.collect()
	torch.cuda.empty_cache()
	import torch

	torch.set_float32_matmul_precision("high")

	from diffusers import HunyuanDiTPipeline
	import argparse
	import time


	def load_pipeline(args):
	"""
	Make sure you have `diffusers`, `accelerate`, `transformers`, and `bitsandbytes` installed.

	You also set up PyTorch and CUDA.

	Once the dependencies are installed, you can run `python run_hunyuan_dit_less_memory.py`.
	"""

	from diffusers import HunyuanDiTPipeline
	from transformers import T5EncoderModel
	from datasets import Dataset, Features
	from datasets import Image as ImageFeature
	from datasets import Value
	import pandas as pd
	import os

	# CSV comes from the notebook above.
	df = pd.read_csv("coco_30k_randomly_sampled_2014_val.csv")
	root_path = "val2014"
	from huggingface_hub import (
	hf_hub_download,
	list_files_info,
	parse_safetensors_file_metadata,
	)
	import json
	import argparse


	def get_files_info(args):
	# SDXL: 0.613, 0.5566, 0.54, 0.4162, 0.4042, 0.4596, 0.5374, 0.5286, 0.5038
	# SD: 0.5396, 0.5707, 0.477, 0.4665, 0.5419, 0.4594, 0.4857, 0.4741, 0.4804

	from diffusers import DiffusionPipeline
	from peft import LoraConfig
	import argparse
	import torch


	def load_pipeline(pipeline_id):