proger

# prompt: https://twitter.com/francoisfleuret/status/1783479122418716805

import os
os.environ['TORCH_LOGS'] = 'output_code' # shows all the bmms
import torch
torch.set_float32_matmul_precision('high')

N, T, D, U, C = 3, 128, 5, 32, 32 # batch, time, heads, head_dim, dim
S = T
A = torch.randn(N, T, D, U) / U**0.5

proger

#%%

import torch
import torch.nn as nn
import torch.nn.functional as F
import matplotlib.pyplot as plt

X = torch.tensor([[0, 0], [0, 1], [1, 0], [1, 1]]).float()
y = torch.logical_xor(X[:, 0], X[:, 1]).float()

proger

proger

"""
# download the model
huggingface-cli download google/gemma-2b-it

# run the server (set the model name here and in the prompt function below)
# notice --kv_cache_dtype fp8_e5m2
docker run --gpus all -p 8000:8000 -e HF_HOME=/hf -e CUDA_VISIBLE_DEVICES=1 -v ~/.cache/huggingface:/hf vllm/vllm-openai:latest:latest --host 0.0.0.0 --model google/gemma-2b-it --kv-cache-dtype fp8_e5m2

# ask one question
echo 'what is python?' | python -m prompt | jq -r '.choices[].text'

proger

import torch


def binary(digits: int) -> torch.IntTensor:
    "Make a basis of powers of two of dimension `digits`, lowest bits to the right"
    return 1 << torch.arange(digits).flip(-1)


def sbt(x: torch.IntTensor) -> torch.IntTensor:
    """Slow Binary Transform.

proger

import argparse
from itertools import islice
import logging
from pathlib import Path

import evaluate
from datasets import load_dataset
from transformers import AutoModelForCausalLM, AutoTokenizer, GenerationConfig
from transformers.generation import BeamSearchDecoderOnlyOutput
from peft import PeftModel

proger

//
// Generated by LLVM NVPTX Back-End
// See scan3 in https://github.com/proger/nanokitchen/blob/main/triscan.py
//

.version 8.1
.target sm_86
.address_size 64

	// .globl	scan3_0d1d2d

proger

import sys
import requests
import json

context = """[INST] They are planning to host a party next weekend. [/INST] Вони планують провести вечірку наступного вікенду.
[INST] I enjoy swimming in the ocean and feeling the salty breeze. [/INST] Мені подобається плавати в океані та відчувати солоний вітер.
[INST]"""

# docker run --gpus all -p 8000:8000 -e HF_HOME=/hf -e CUDA_VISIBLE_DEVICES=1 -v ~/.cache/huggingface:/hf ghcr.io/mistralai/mistral-src/vllm:latest --host 0.0.0.0 --model mistralai/Mistral-7B-v0.1
def prompt(input, url="http://localhost:8000/v1/completions"):

proger

import torch
from torch import nn
from torch.utils.data import DataLoader, TensorDataset
import gzip                                                                                                    
import numpy as np                                                                                             

torch.set_float32_matmul_precision('high')


def read(filename):                                    

proger

import gzip
import numpy as np
from sklearn.cluster import MiniBatchKMeans

def read(filename):
    with gzip.open(filename, 'rb') as file:
        compressed_data = file.read()

    data = np.frombuffer(compressed_data, dtype=np.float32)
    return data