Skip to content

Instantly share code, notes, and snippets.

@DOUDOU0314
DOUDOU0314 / generation_example.py
Created June 18, 2021 07:52
generation example of GPT-J-6B: 6B JAX-Based Transformer
import time
import torch
from transformers import GPTNeoForCausalLM, AutoConfig, GPT2Tokenizer
import torch
import transformers
import collections
import os
@DOUDOU0314
DOUDOU0314 / comparison.txt
Last active January 11, 2021 07:51
Comparison of deepspeed and gluon-nlp in SQuAD1.1
Comparison of deepspeed and gluon-nlp in SQuAD1.1
Dataset:SQuAD1.1
GPUs:p3.8x with four teslaV100
BatchSize: 3
Epoch: 2
Max_seq_len: 384
Model: google_en_uncased_bert_wwm_large
Results:
deepspeed : 93.11/87.03 time cost: 0.71hours