T5 Base Qa Summary Emotion

🚀 T5 Base 模型：问答 + 摘要 + 情感分析

本模型融合了问答、文本摘要和情感检测功能，在多个权威数据集上进行微调训练，能为文本处理任务提供高效准确的解决方案。

🚀 快速开始

依赖项

需要 transformers>=4.0.0

✨ 主要特性

• 多任务支持：支持问答、文本摘要和情感检测三种任务。
• 微调训练：在 CoQa、Squad 2、GoEmotions 和 CNN/DailyMail 数据集上进行微调。
• 优秀表现：在 Squad 2 开发集上达到 F1 79.5 的分数，在 CoQa 开发集上达到 F1 70.6 的分数。

📦 安装指南

确保安装 transformers 库，版本需大于等于 4.0.0：

pip install transformers>=4.0.0

💻 使用示例

基础用法

问答任务

使用 Transformers 库

from transformers import T5ForConditionalGeneration, T5Tokenizer
model = T5ForConditionalGeneration.from_pretrained("kiri-ai/t5-base-qa-summary-emotion")
tokenizer = T5Tokenizer.from_pretrained("kiri-ai/t5-base-qa-summary-emotion")

def get_answer(question, prev_qa, context):
    input_text = [f"q: {qa[0]} a: {qa[1]}" for qa in prev_qa]
    input_text.append(f"q: {question}")
    input_text.append(f"c: {context}")
    input_text = " ".join(input_text)
    features = tokenizer([input_text], return_tensors='pt')
    tokens = model.generate(input_ids=features['input_ids'], 
            attention_mask=features['attention_mask'], max_length=64)
    return tokenizer.decode(tokens[0], skip_special_tokens=True)

print(get_answer("Why is the moon yellow?", "I'm not entirely sure why the moon is yellow.")) # unknown

context = "Elon Musk left OpenAI to avoid possible future conflicts with his role as CEO of Tesla."

print(get_answer("Why not?", [("Does Elon Musk still work with OpenAI", "No")], context)) # to avoid possible future conflicts with his role as CEO of Tesla

使用 Kiri 库

from kiri.models import T5QASummaryEmotion

context = "Elon Musk left OpenAI to avoid possible future conflicts with his role as CEO of Tesla."
prev_qa = [("Does Elon Musk still work with OpenAI", "No")]
model = T5QASummaryEmotion()

# Leave prev_qa blank for non conversational question-answering
model.qa("Why not?", context, prev_qa=prev_qa)
> "to avoid possible future conflicts with his role as CEO of Tesla"

文本摘要任务

使用 Transformers 库

from transformers import T5ForConditionalGeneration, T5Tokenizer
model = T5ForConditionalGeneration.from_pretrained("kiri-ai/t5-base-qa-summary-emotion")
tokenizer = T5Tokenizer.from_pretrained("kiri-ai/t5-base-qa-summary-emotion")

def summary(context):
    input_text = f"summarize: {context}"
    features = tokenizer([input_text], return_tensors='pt')
    tokens = model.generate(input_ids=features['input_ids'], 
            attention_mask=features['attention_mask'], max_length=64)
    return tokenizer.decode(tokens[0], skip_special_tokens=True)

使用 Kiri 库

from kiri.models import T5QASummaryEmotion

model = T5QASummaryEmotion()

model.summarise("Long text to summarise")
> "Short summary of long text"

情感检测任务

使用 Transformers 库

from transformers import T5ForConditionalGeneration, T5Tokenizer
model = T5ForConditionalGeneration.from_pretrained("kiri-ai/t5-base-qa-summary-emotion")
tokenizer = T5Tokenizer.from_pretrained("kiri-ai/t5-base-qa-summary-emotion")

def emotion(context):
    input_text = f"emotion: {context}"
    features = tokenizer([input_text], return_tensors='pt')
    tokens = model.generate(input_ids=features['input_ids'], 
            attention_mask=features['attention_mask'], max_length=64)
    return tokenizer.decode(tokens[0], skip_special_tokens=True)

使用 Kiri 库

from kiri.models import T5QASummaryEmotion

model = T5QASummaryEmotion()

model.emotion("I hope this works!")
> "optimism"

📚 详细文档

描述

该模型在 CoQa、Squad 2、GoEmotions 和 CNN/DailyMail 数据集上进行了微调。

在 Squad 2 开发集上达到了 F1 79.5 的分数，在 CoQa 开发集上达到了 F1 70.6 的分数。

文本摘要和情感检测功能尚未进行评估。

📄 许可证

本项目采用 Apache-2.0 许可证。

关于我们

Kiri 让使用最先进的模型变得简单、便捷且可扩展。

官网 | 自然语言引擎

📦 相关信息