tiktoken-rust 0.2.1

a fast BPE tokeniser for use with OpenAI's models
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39

import base64
import functools
import gzip
import json
import os
import random
import time
from typing import Any, cast

import blobfile

import tiktoken


def benchmark_batch(documents: list[str]) -> None:
    num_threads = int(os.environ["RAYON_NUM_THREADS"])
    num_bytes = sum(map(len, map(str.encode, documents)))
    print(f"num_threads: {num_threads}, num_bytes: {num_bytes}")

    enc = tiktoken.get_encoding("gpt2")
    enc.encode("warmup")

    start = time.perf_counter_ns()
    enc.encode_ordinary_batch(documents, num_threads=num_threads)
    end = time.perf_counter_ns()
    print(f"tiktoken \t{num_bytes / (end - start) * 1e9} bytes / s")

    import transformers

    hf_enc = cast(Any, transformers).GPT2TokenizerFast.from_pretrained("gpt2")
    hf_enc.model_max_length = 1e30  # silence!
    hf_enc.encode("warmup")

    start = time.perf_counter_ns()
    hf_enc(documents)
    end = time.perf_counter_ns()
    print(f"huggingface \t{num_bytes / (end - start) * 1e9} bytes / s")