ВУВ (Виртуальный Ускоритель Вычислений) - ускоряет работу и скорость обучения трансформерных нейросетей в x2-x6 раз (GPT2 - в 40 раз)
Project description
Пример использования , на примере GPT 2-й версии :
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained(
"gpt2",
quantization_bits=8, #биты для квантования
sparsity_threshold=0.02 #обнуление малых весов (0.01 - 0.05)
)
В разы ускоряет вычисления нейронных сетей (в основном рассматривается как для трансформеров)
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
eridaspeed-0.1.0.tar.gz
(2.6 kB
view details)
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file eridaspeed-0.1.0.tar.gz.
File metadata
- Download URL: eridaspeed-0.1.0.tar.gz
- Upload date:
- Size: 2.6 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.1.0 CPython/3.12.3
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
8c8cec061435abfd0de034067bbe6b1a578342266ee61f23795bb5cb2cdee917
|
|
| MD5 |
9f7cb86cbdaa2db4309385bea33e002b
|
|
| BLAKE2b-256 |
35a00afd6af51700a16386e3b29e229125d5c8a6f7736d47b550dbd08e8eea16
|
File details
Details for the file eridaspeed-0.1.0-py3-none-any.whl.
File metadata
- Download URL: eridaspeed-0.1.0-py3-none-any.whl
- Upload date:
- Size: 2.8 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.1.0 CPython/3.12.3
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
3ba665aeb31aff97cd437a5fd1efb751b89bc96579c46f0ed118f2157ac7e060
|
|
| MD5 |
92cb9be16747979f520badfbf3f8fae4
|
|
| BLAKE2b-256 |
7da0cce9bf879ca3d62ebd408251edeef36989b96ee7e305b276c46e915d5b26
|
