distilbert base uncased finetuned sst 2 english

提供商distilbert
分类text-classification
许可证apache-2.0
下载量2.3K
星标1

简介

这是一个基于 DistilBERT 的轻量级文本分类模型，专门在 SST-2 情感分析数据集上进行了微调。简单来说，它是一个高效的“情感检测器”，能快速判断英文文本是正面还是负面评价。相比于原版 BERT，它在保持绝大部分性能的同时，显著降低了推理延迟和内存占用，非常适合部署在资源有限的端侧设备或需要高吞吐量的实时分析场景中。对于开发者而言，它无需复杂配置即可快速上手，是构建英文评论分析、用户反馈筛选等功能的理想基准模型。

核心亮点

  • 轻量化架构，推理速度快且内存占用低
  • 专攻英文情感分析，精准识别正负面情绪
  • Apache-2.0 协议，商业部署无压力
  • 低上手门槛，适合作为文本分类的快速原型

使用方法

安装依赖
# 安装 Hugging Face transformers
pip install transformers torch
SDK 使用
# 使用 transformers 加载模型
from transformers import AutoModel, AutoTokenizer

model = AutoModel.from_pretrained("distilbert/distilbert-base-uncased-finetuned-sst-2-english")
tokenizer = AutoTokenizer.from_pretrained("distilbert/distilbert-base-uncased-finetuned-sst-2-english")

Hugging Face 下载

我们推荐使用命令行或者 Hugging Face Hub SDK 来进行模型的下载。

操作指引：在下载前，请先通过如下命令安装 huggingface_hub：

操作指引 
pip install -U huggingface_hub

命令行下载

下载完整模型库

下载完整模型库 
huggingface-cli download distilbert/distilbert-base-uncased-finetuned-sst-2-english

下载单个文件到指定本地文件夹（以下载 config.json 到当前路径下 ./dir 目录为例）

下载单个文件到指定本地文件夹（以下载 config.json 到当前路径下 ./dir 目录为例） 
huggingface-cli download distilbert/distilbert-base-uncased-finetuned-sst-2-english config.json --local-dir ./dir

更多命令行下载选项，可参见官方文档

SDK 下载

SDK 下载 
# 模型下载
from huggingface_hub import snapshot_download
model_dir = snapshot_download('distilbert/distilbert-base-uncased-finetuned-sst-2-english')

Git 下载

请确保 lfs 已经被正确安装

Git 下载 
git lfs install
git clone https://huggingface.co/distilbert/distilbert-base-uncased-finetuned-sst-2-english

如果您希望跳过 lfs 大文件下载，可以使用如下命令

跳过 LFS 
GIT_LFS_SKIP_SMUDGE=1 git clone https://huggingface.co/distilbert/distilbert-base-uncased-finetuned-sst-2-english

模型文件托管在 Hugging Face Hub，使用 HF CLI / SDK / Git 直接下载，不经过本站。

PyTorch / Transformers 使用

安装 Transformers

安装 Transformers 
pip install -U transformers torch

模型加载和推理

模型加载和推理 
from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained('distilbert/distilbert-base-uncased-finetuned-sst-2-english')
tokenizer = AutoTokenizer.from_pretrained('distilbert/distilbert-base-uncased-finetuned-sst-2-english')

模型下载

我们推荐使用命令行或者 ModelScope SDK 来进行模型的下载。

操作指引：在下载前，请先通过如下命令安装 ModelScope：

操作指引 
pip install modelscope

命令行下载

下载完整模型库

下载完整模型库 
modelscope download --model distilbert/distilbert-base-uncased-finetuned-sst-2-english

下载单个文件到指定本地文件夹（以下载 README.md 到当前路径下 dir 目录为例）

下载单个文件到指定本地文件夹（以下载 README.md 到当前路径下 dir 目录为例） 
modelscope download --model distilbert/distilbert-base-uncased-finetuned-sst-2-english README.md --local_dir ./dir

更多更丰富的命令行下载选项，可参见具体文档

SDK 下载

SDK 下载 
# 模型下载
from modelscope import snapshot_download
model_dir = snapshot_download('distilbert/distilbert-base-uncased-finetuned-sst-2-english')

Git 下载

请确保 lfs 已经被正确安装

Git 下载 
git lfs install
git clone https://www.modelscope.cn/distilbert/distilbert-base-uncased-finetuned-sst-2-english.git

如果您希望跳过 lfs 大文件下载，可以使用如下命令

跳过 LFS 
GIT_LFS_SKIP_SMUDGE=1 git clone https://www.modelscope.cn/distilbert/distilbert-base-uncased-finetuned-sst-2-english.git

ModelScope 模型页直接下载模型文件；无需将模型文件放在本站服务器。

Notebook 快速开发

下载并安装 ModelScope library

下载并安装 ModelScope library 
pip install "modelscope[audio,cv,nlp,multi-modal,science]" -f https://modelscope.oss-cn-beijing.aliyuncs.com/releases/repo.html

模型加载和推理

模型加载和推理 
from modelscope.pipelines import pipeline
from modelscope.utils.constant import Tasks

p = pipeline('text-generation', 'distilbert/distilbert-base-uncased-finetuned-sst-2-english')

完整文档

来源: HuggingFace

---
language: en
license: apache-2.0
datasets:

  • sst2

  • glue

model-index:
  • name: distilbert-base-uncased-finetuned-sst-2-english

results:
- task:
type: text-classification
name: Text Classification
dataset:
name: glue
type: glue
config: sst2
split: validation
metrics:
- type: accuracy
value: 0.9105504587155964
name: Accuracy
verified: true
verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiN2YyOGMxYjY2Y2JhMjkxNjIzN2FmMjNiNmM2ZWViNGY3MTNmNWI2YzhiYjYxZTY0ZGUyN2M1NGIxZjRiMjQwZiIsInZlcnNpb24iOjF9.uui0srxV5ZHRhxbYN6082EZdwpnBgubPJ5R2-Wk8HTWqmxYE3QHidevR9LLAhidqGw6Ih93fK0goAXncld_gBg
- type: precision
value: 0.8978260869565218
name: Precision
verified: true
verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiMzgwYTYwYjA2MmM0ZTYwNDk0M2NmNTBkZmM2NGNhYzQ1OGEyN2NkNDQ3Mzc2NTQyMmZiNDJiNzBhNGVhZGUyOSIsInZlcnNpb24iOjF9.eHjLmw3K02OU69R2Au8eyuSqT3aBDHgZCn8jSzE3_urD6EUSSsLxUpiAYR4BGLD_U6-ZKcdxVo_A2rdXqvUJDA
- type: recall
value: 0.9301801801801802
name: Recall
verified: true
verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiMGIzM2E3MTI2Mzc2MDYwNmU3ZTVjYmZmZDBkNjY4ZTc5MGY0Y2FkNDU3NjY1MmVkNmE3Y2QzMzAwZDZhOWY1NiIsInZlcnNpb24iOjF9.PUZlqmct13-rJWBXdHm5tdkXgETL9F82GNbbSR4hI8MB-v39KrK59cqzFC2Ac7kJe_DtOeUyosj34O_mFt_1DQ
- type: auc
value: 0.9716626673402374
name: AUC
verified: true
verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiMDM0YWIwZmQ4YjUwOGZmMWU2MjI1YjIxZGQ2MzNjMzRmZmYxMzZkNGFjODhlMDcyZDM1Y2RkMWZlOWQ0MWYwNSIsInZlcnNpb24iOjF9.E7GRlAXmmpEkTHlXheVkuL1W4WNjv4JO3qY_WCVsTVKiO7bUu0UVjPIyQ6g-J1OxsfqZmW3Leli1wY8vPBNNCQ
- type: f1
value: 0.9137168141592922
name: F1
verified: true
verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiMGU4MjNmOGYwZjZjMDQ1ZTkyZTA4YTc1MWYwOTM0NDM4ZWY1ZGVkNDY5MzNhYTQyZGFlNzIyZmUwMDg3NDU0NyIsInZlcnNpb24iOjF9.mW5ftkq50Se58M-jm6a2Pu93QeKa3MfV7xcBwvG3PSB_KNJxZWTCpfMQp-Cmx_EMlmI2siKOyd8akYjJUrzJCA
- type: loss
value: 0.39013850688934326
name: loss
verified: true
verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiMTZiNzAyZDc0MzUzMmE1MGJiN2JlYzFiODE5ZTNlNGE4MmI4YzRiMTc2ODEzMTUwZmEzOTgxNzc4YjJjZTRmNiIsInZlcnNpb24iOjF9.VqIC7uYC-ZZ8ss9zQOlRV39YVOOLc5R36sIzCcVz8lolh61ux_5djm2XjpP6ARc6KqEnXC4ZtfNXsX2HZfrtCQ
- task:
type: text-classification
name: Text Classification
dataset:
name: sst2
type: sst2
config: default
split: train
metrics:
- type: accuracy
value: 0.9885521685548412
name: Accuracy
verified: true
verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiY2I3NzU3YzhmMDkxZTViY2M3OTY1NmI0ZTdmMDQxNjNjYzJiZmQxNzczM2E4YmExYTY5ODY0NDBkY2I4ZjNkOCIsInZlcnNpb24iOjF9.4Gtk3FeVc9sPWSqZIaeUXJ9oVlPzm-NmujnWpK2y5s1Vhp1l6Y1pK5_78wW0-NxSvQqV6qd5KQf_OAEpVAkQDA
- type: precision
value: 0.9881965062029833
name: Precision Macro
verified: true
verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiZDdlZDMzY2I3MTAwYTljNmM4MGMyMzU2YjAzZDg1NDYwN2ZmM2Y5OWZhMjUyMGJiNjY1YmZiMzFhMDI2ODFhNyIsInZlcnNpb24iOjF9.cqmv6yBxu4St2mykRWrZ07tDsiSLdtLTz2hbqQ7Gm1rMzq9tdlkZ8MyJRxtME_Y8UaOG9rs68pV-gKVUs8wABw
- type: precision
value: 0.9885521685548412
name: Precision Micro
verified: true
verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiZjFlYzAzNmE1YjljNjUwNzBjZjEzZDY0ZDQyMmY5ZWM2OTBhNzNjYjYzYTk1YWE1NjU3YTMxZDQwOTE1Y2FkNyIsInZlcnNpb24iOjF9.jnCHOkUHuAOZZ_ZMVOnetx__OVJCS6LOno4caWECAmfrUaIPnPNV9iJ6izRO3sqkHRmxYpWBb-27GJ4N3LU-BQ
- type: precision
value: 0.9885639626373408
name: Precision Weighted
verified: true
verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiZGUyODFjNjBlNTE2MTY3ZDAxOGU1N2U0YjUyY2NiZjhkOGVmYThjYjBkNGU3NTRkYzkzNDQ2MmMwMjkwMWNiMyIsInZlcnNpb24iOjF9.zTNabMwApiZyXdr76QUn7WgGB7D7lP-iqS3bn35piqVTNsv3wnKjZOaKFVLIUvtBXq4gKw7N2oWxvWc4OcSNDg
- type: recall
value: 0.9886145346602994
name: Recall Macro
verified: true
verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiNTU1YjlhODU3YTkyNTdiZDcwZGFlZDBiYjY0N2NjMGM2NTRiNjQ3MDNjNGMxOWY2ZGQ4NWU1YmMzY2UwZTI3YSIsInZlcnNpb24iOjF9.xaLPY7U-wHsJ3DDui1yyyM-xWjL0Jz5puRThy7fczal9x05eKEQ9s0a_WD-iLmapvJs0caXpV70hDe2NLcs-DA
- type: recall
value: 0.9885521685548412
name: Recall Micro
verified: true
verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiODE0YTU0MDBlOGY4YzU0MjY5MzA3OTk2OGNhOGVkMmU5OGRjZmFiZWI2ZjY5ODEzZTQzMTI0N2NiOTVkNDliYiIsInZlcnNpb24iOjF9.SOt1baTBbuZRrsvGcak2sUwoTrQzmNCbyV2m1_yjGsU48SBH0NcKXicidNBSnJ6ihM5jf_Lv_B5_eOBkLfNWDQ
- type: recall
value: 0.9885521685548412
name: Recall Weighted
verified: true
verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiZWNkNmM0ZGRlNmYxYzIwNDk4OTI5MzIwZWU1NzZjZDVhMDcyNDFlMjBhNDQxODU5OWMwMWNhNGEzNjY3ZGUyOSIsInZlcnNpb24iOjF9.b15Fh70GwtlG3cSqPW-8VEZT2oy0CtgvgEOtWiYonOovjkIQ4RSLFVzVG-YfslaIyfg9RzMWzjhLnMY7Bpn2Aw
- type: f1
value: 0.9884019815052447
name: F1 Macro
verified: true
verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiYmM4NjQ5Yjk5ODRhYTU1MTY3MmRhZDBmODM1NTg3OTFiNWM4NDRmYjI0MzZkNmQ1MzE3MzcxODZlYzBkYTMyYSIsInZlcnNpb24iOjF9.74RaDK8nBVuGRl2Se_-hwQvP6c4lvVxGHpcCWB4uZUCf2_HoC9NT9u7P3pMJfH_tK2cpV7U3VWGgSDhQDi-UBQ
- type: f1
value: 0.9885521685548412
name: F1 Micro
verified: true
verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiZDRmYWRmMmQ0YjViZmQxMzhhYTUyOTE1MTc0ZDU1ZjQyZjFhMDYzYzMzZDE0NzZlYzQyOTBhMTBhNmM5NTlkMiIsInZlcnNpb24iOjF9.VMn_psdAHIZTlW6GbjERZDe8MHhwzJ0rbjV_VJyuMrsdOh5QDmko-wEvaBWNEdT0cEKsbggm-6jd3Gh81PfHAQ
- type: f1
value: 0.9885546181087554
name: F1 Weighted
verified: true
verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiMjUyZWFhZDZhMGQ3MzBmYmRiNDVmN2FkZDBjMjk3ODk0OTAxNGZkMWE0NzU5ZjI0NzE0NGZiNzM0N2Y2NDYyOSIsInZlcnNpb24iOjF9.YsXBhnzEEFEW6jw3mQlFUuIrW7Gabad2Ils-iunYJr-myg0heF8NEnEWABKFE1SnvCWt-69jkLza6SupeyLVCA
- type: loss
value: 0.040652573108673096
name: loss
verified: true
verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiZTc3YjU3MjdjMzkxODA5MjU5NGUyY2NkMGVhZDg3ZWEzMmU1YWVjMmI0NmU2OWEyZTkzMTVjNDZiYTc0YjIyNCIsInZlcnNpb24iOjF9.lA90qXZVYiILHMFlr6t6H81Oe8a-4KmeX-vyCC1BDia2ofudegv6Vb46-4RzmbtuKeV6yy6YNNXxXxqVak1pAg
---

DistilBERT base uncased finetuned SST-2

Table of Contents

Model Details

Model Description: This model is a fine-tune checkpoint of DistilBERT-base-uncased, fine-tuned on SST-2. This model reaches an accuracy of 91.3 on the dev set (for comparison, Bert bert-base-uncased version reaches an accuracy of 92.7).
  • Developed by: Hugging Face
  • Model Type: Text Classification
  • Language(s): English
  • License: Apache-2.0
  • Parent Model: For more details about DistilBERT, we encourage users to check out this model card.
  • Resources for more information:
- Model Documentation - DistilBERT paper

How to Get Started With the Model

Example of single-label classification:
​​
```python
import torch
from transformers import DistilBertTokenizer, DistilBertForSequenceClassification

tokenizer = DistilBertTokenizer.from_pretrained("distilbert-base-uncased-finetuned-sst-2-english")
model = DistilBertForSequenceClassification.from_pretrained("distilbert-base-uncased-fi

查看官方来源