| import argparse | |
| import random | |
| import sys | |
| from transformers import AutoModelForCausalLM, AutoTokenizer, DynamicCache | |
| import torch | |
| parser = argparse.ArgumentParser() | |
| parser.add_argument("question", type=str) | |
| parser.add_argument( |
| # Clone llama.cpp | |
| git clone https://github.com/ggerganov/llama.cpp.git | |
| cd llama.cpp | |
| # Build it | |
| make clean | |
| LLAMA_METAL=1 make | |
| # Download model | |
| export MODEL=llama-2-13b-chat.ggmlv3.q4_0.bin |
The screenshots were taken on different sessions.
The entire sessions are included on the screenshots.
I lost the original prompts, so I had to reconstruct them, and still managed to reproduce.
The "compressed" version is actually longer! Emojis and abbreviations use more tokens than common words.
| This is free and unencumbered software released into the public domain. | |
| Anyone is free to copy, modify, publish, use, compile, sell, or | |
| distribute this software, either in source code form or as a compiled | |
| binary, for any purpose, commercial or non-commercial, and by any | |
| means. | |
| In jurisdictions that recognize copyright laws, the author or authors | |
| of this software dedicate any and all copyright interest in the | |
| software to the public domain. We make this dedication for the benefit |
Have you ever tried to merge two branches only to end up in conflict hell? You fix a bunch of conflicts only to run git merge --continue and be presented with the same conflicts. Repeat this process and after a few iterations you give up because it just isn't worth the pain and effort.
Would you be surprised to know that there is a git feature specifically for this problem? It's called rerere and I'm going to enrich your life with it now. (I'm going to talk specifically about merging but I think it also helps rebasing)
rerere stands for Reuse Recorded Resolution. The TL;DR version is you ask git to remember how you've resolved hunks in the past, and if the same one comes up for a file in future just redo what you did last time.
To enable this feature just run this lovely command git config --global rerere.enabled true. You can also turn it on by creating this directory in your projects .git/rr-cache, although the global setting is much clearer.
I'll try to take you through an example of how th
| from alembic import config | |
| from alembic import script | |
| from alembic.runtime import migration | |
| import sqlalchemy | |
| import exceptions | |
| engine = sqlalchemy.create_engine(DATABASE_URL) | |
| alembic_cfg = config.Config('alembic.ini') |
Public
Dizzying but invisible depth
You just went to the Google home page.
| from selenium import webdriver | |
| from selenium.webdriver.chrome.options import Options | |
| chrome_options = Options() | |
| chrome_options.add_argument("--headless") | |
| chrome_options.binary_location = '/Applications/Google Chrome.app/Contents/MacOS/Google Chrome' | |
| driver = webdriver.Chrome(executable_path='/usr/local/bin/chromedriver', chrome_options=chrome_options) | |
| driver.get("http://www.duo.com") |
| import numpy as np | |
| from sklearn import datasets | |
| from sklearn.semi_supervised import LabelPropagation, LabelSpreading | |
| ###for n_samples in [20, 200, 2000, 20000]: | |
| ### X, y = datasets.make_classification(n_samples=n_samples, n_classes=3, n_informative=3) | |
| for (X, y) in [datasets.load_iris(return_X_y=True)]: | |
| for model in [LabelPropagation(max_iter=1000), | |
| #LabelSpreading(alpha=0.01), | |
| #LabelSpreading(alpha=0.1), | |
| #LabelSpreading(alpha=0.3) |
| #!/bin/bash | |
| set -e | |
| set -u | |
| set -o pipefail | |
| if [ -z $1 ] | |
| then | |
| echo "Usage: $0 [user@]servername" | |
| exit 1 |
