cloneofsimo

#define DELEGATE_FAST_GELU_GRADIENT_CUDA_KERNEL(T, FMAFunc, TanhFunc)    \
  template <>                                                            \
  __global__ void FastGeluGradientCUDAKernel<T>(                         \
      const int N, const T* dY, const T* X, T* dX) {                     \
    constexpr T kAlpha = M_2_SQRTPI * M_SQRT1_2;                         \
    constexpr T kBeta = kAlpha * gelu_utils::kFastCoeff * T(3);          \
    const int index = blockIdx.x * CAFFE_CUDA_NUM_THREADS + threadIdx.x; \
    if (index < N) {                                                     \
      const T y = TanhFunc(                                              \
          kAlpha *                                                       \

cloneofsimo

import pytest
import torch

import triton
import triton.language as tl


@triton.jit
def _fwd_kernel(
    Q, K, V, sm_scale,

cloneofsimo

from diffusers import DiffusionPipeline
from typing import Any, Callable, Dict, List, Optional, Tuple, Union
import torch
import re

MODEL_CACHE = "./cache"


def split_by_emphasis(text, tokenizer, normalize = True):
    

cloneofsimo

# Bootstrapped from Huggingface diffuser's code.

import gc
import math
from pathlib import Path
import torch
import torch.nn as nn

import torch.nn.functional as F

cloneofsimo

import torch
from diffusers import DiffusionPipeline

from safetensors import safe_open
from dataset_and_utils import TokenEmbeddingsHandler


MODEL_CACHE = "./cache"
pipe = DiffusionPipeline.from_pretrained(

cloneofsimo

 from rg2.gym import Rg2UEnv, WalkerEnvConfig
 
 from gym.wrappers import TimeLimit
 from stable_baselines3 import PPO, SAC
 from stable_baselines3.common.monitor import Monitor
 from stable_baselines3.common.vec_env import VecNormalize, DummyVecEnv, SubprocVecEnv
 from stable_baselines3.common.callbacks import CheckpointCallback
 
 
 if __name__ == "__main__":

cloneofsimo

import os
import csv
import torch
import cv2
import logging
from typing import Tuple, Any, List
from torch.utils.data import DataLoader, Dataset
from multiprocessing import Pool
from streaming import MDSWriter
import ImageReward as RM

cloneofsimo

import torch
import torch.nn.functional as F
from PIL import Image
import matplotlib.pyplot as plt
import numpy as np
import torch.nn as nn

# Load an example image
image_path = "/home/simo/just_dl_stuff/vid2data/vender_1.png"
img = Image.open(image_path).convert("RGB")

cloneofsimo

clean clean

black .
autoflake --remove-all-unused-imports -i ./*
isort .

Kill cuda shit

cloneofsimo

Setting up a Docker environment with PyTorch on your Linux server so that your friends can access it involves several steps, including installing Docker, pulling or building a PyTorch Docker image, and configuring Docker to ensure it is accessible securely from other machines. Here’s a step-by-step guide to get you started:

Step 1: Install Docker

1. Update your system: Ensure your package lists and installed packages are updated.

   sudo apt update && sudo apt upgrade -y

2. Install Docker: Install Docker using the convenience script from Docker which works on many Linux distributions.