For example, kill command contains python3 -u experiment_main.py
kill $(ps aux | grep '[p]ython3 -u experiment_main.py' | awk '{print $2}')
hdfs dfs -ls / | sort -k6,7
pip install streamlit | |
pip install spacy | |
python -m spacy download en_core_web_sm | |
python -m spacy download en_core_web_md | |
python -m spacy download de_core_news_sm |
<!DOCTYPE html> | |
<head> | |
<!--Little CSS fade in --> | |
<style> | |
.fade-in{ | |
-webkit-animation: fade-in 2s ease; | |
-moz-animation: fade-in ease-in-out 2s both; | |
-ms-animation: fade-in ease-in-out 2s both; | |
-o-animation: fade-in ease-in-out 2s both; |
import torch | |
from fairseq.models.bart import BARTModel | |
bart = BARTModel.from_pretrained( | |
'model_files/bart-large-model', | |
checkpoint_file='checkpoint_best.pt', | |
data_name_or_path='data/cloze_replace_all-bin' | |
) | |
bart.cuda() |
from torch.utils.data import DataLoader | |
from transformers import AutoTokenizer, PreTrainedTokenizerFast, set_seed, AutoModelForCausalLM, AutoConfig | |
from tqdm import tqdm | |
import argparse | |
import torch | |
import torch.nn as nn | |
import logging | |
from typing import Dict, Tuple | |
from accelerate import Accelerator, DistributedDataParallelKwargs | |
from accelerate.logging import get_logger |
""" | |
This is a simple example to show how to calculate the p_value of two models' accuracy | |
Bootstrapint t-test | |
""" | |
import random | |
random.seed(42) | |
# assume we have test set 1000 samples | |
# we just create dummy results to demo | |
groundtruth = [random.choice(['A', 'B', 'C']) for _ in range(1000)] |
{"name": "HumanEval_79_decimal_to_binary", "language": "py", "prompt": "def decimal_to_binary(decimal: int) -> str:\n \"\"\"You will be given a number in decimal form and your task is to convert it to\n binary format. The function should return a string, with each character representing a binary\n number. Each character in the string will be '0' or '1'.\n\n There will be an extra couple of characters 'db' at the beginning and at the end of the string.\n The extra characters are there to help with the format.\n\n Examples:\n >>> decimal_to_binary(15)\n 'db1111db'\n >>> decimal_to_binary(32)\n 'db100000db'\n \"\"\"\n", "doctests": "transform", "original": "/home/arjun/repos/nuprl/MultiPL-E/datasets/../datasets/originals-with-cleaned-doctests/HumanEval_79_decimal_to_binary.py", "prompt_terminology": "reworded", "stop_tokens": ["\ndef", "\n#", "\nif", "\nclass"], "entry_point": "decimal_to_binary", "test": "def check(candidate):\n assert candidate(0) == 'db0db'\n assert cand |
import openai | |
import asyncio | |
async def get_choice_completion(prompt, choices): | |
# Initialize an asynchronous OpenAI client | |
async with openai.AsyncClient(base_url="http://127.0.0.1:8000/v1", api_key="abc") as client: | |
choice_probs = {} | |
# Calculate logprobs for each prompt + choice sequence | |
for choice in choices: |
MATH_PROMPT = '''Question: Riku has 25 times more stickers than Kristoff. If Kristoff has 85 stickers, how many stickers does Riku have? | |
Answer: | |
If Kristoff has 85 stickers, Riku has 25 * 85 stickers = <<85*25=2125>>2125 more stickers. | |
The total number of stickers Riku has is 2125 stickers + 85 stickers = <<2125+85=2210>>2210 stickers. | |
# solution in Python: | |
def solution(): | |
"""Riku has 25 times more stickers than Kristoff. If Kristoff has 85 stickers, how many stickers does Riku have?""" | |
ratio = 25 |
PYTHON_PROMPT = '''Question: after resting they decided to go for a swim . the depth of the water is 5 times ron 's height . dean is 11 feet shorter than ron . if ron stands at 12 feet how deep was the water ? | |
# solution in Python: | |
def solution(): | |
"""after resting they decided to go for a swim . if the depth of the water is 10 times dean 's height and he stands at 6 feet how deep was the water ?""" | |
dean_height = 6 | |
water_depth = 10 * dean_height | |
result = water_depth | |
return result |