#jupyter_notebook#chinese_llm#chinese_nlp#finetune#generative_ai#instruct_gpt#instruction_set#llama#llm#lora#open_models#open_source#open_source_models#qlora
AirLLM is a tool that lets you run very large AI models on computers with limited memory by using a smart layer-by-layer loading technique instead of traditional compression methods. You can run a 70-billion-parameter model on just 4GB of GPU memory, or even a 405-billion-parameter model on 8GB, without losing model quality. The benefit is that you can use powerful AI models on affordable hardware without expensive upgrades, and the tool also offers optional compression features that can speed up performance by up to 3 times while maintaining accuracy.
https://github.com/lyogavin/airllm
# The standard string repr for dicts is hard to read:
»> my_mapping = {'a': 23, 'b': 42, 'c': 0xc0ffee}
»> my_mapping
{'b': 42, 'c': 12648430. 'a': 23} # 😞
# The "#json" module can do a much better job:
»> import json
»> print(json.dumps(my_mapping, indent=4, sort_keys=True))
{
"a": 23,
"b": 42,
"c": 12648430
}
# Note this only works with dicts containing
# primitive types (check out the "pprint" module):
»> json.dumps({all: 'yup'})
TypeError: keys must be a string
История(12м) как в Альфа-Банке сокращали размер JSON файла, который передается на устройство для работы SDUI. Решением стала шаблонизация для отказа от одинаковых блоков UI с разными данными
#оптимизация#json
¿Que puede hacer este bot?
@apimaniaBot
Con éste bot puedes crear PDF a partir de páginas web, convertir texto a imágenes, convertir tablas HTML a json y mucho más
Idioma: español
(Visto en @botsgram_cu)
#pdf#web#texto#imágenes#hrml#json