pip install transformers
{
"attention_probs_dropout_prob": 0.1,
"hidden_act": "gelu",
"hidden_dropout_prob": 0.1,
"hidden_size": 768,
"initializer_range": 0.02,
"intermediate_size": 3072,
"max_position_embeddings": 512,
"num_attention_heads": 12,
"num_hidden_layers": 12,
"type_vocab_size": 2,
"vocab_size": 28996
}
from transformers import (TFBertModel, BertTokenizer,
TFGPT2Model, GPT2Tokenizer)
bert_model = TFBertModel.from_pretrained("bert-base-cased") # Automatically loads the config
bert_tokenizer = BertTokenizer.from_pretrained("bert-base-cased")
gpt2_model = TFGPT2Model.from_pretrained("gpt2") # Automatically loads the config
gpt2_tokenizer = GPT2Tokenizer.from_pretrained("gpt2")
import tensorflow_datasets
from transformers import glue_convert_examples_to_features
data = tensorflow_datasets.load("glue/mrpc")
train_dataset = data["train"]
validation_dataset = data["validation"]
train_dataset = glue_convert_examples_to_features(train_dataset, bert_tokenizer, 128, 'mrpc')
validation_dataset = glue_convert_examples_to_features(validation_dataset, bert_tokenizer, 128, 'mrpc')
train_dataset = train_dataset.shuffle(100).batch(32).repeat(2)
validation_dataset = validation_dataset.batch(64)
optimizer = tf.keras.optimizers.Adam(learning_rate=3e-5, epsilon=1e-08, clipnorm=1.0)
loss = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True)
metric = tf.keras.metrics.SparseCategoricalAccuracy('accuracy')
bert_model.compile(optimizer=optimizer, loss=loss, metrics=[metric])
bert_history = bert_model.fit(
bert_train_dataset,
epochs=2,
steps_per_epoch=115,
validation_data=bert_validation_dataset,
validation_steps=7
)
model = TFBertForSequenceClassification.from_pretrained("bert-base-cased")
tokenizer = BertTokenizer.from_pretrained("bert-base-cased")
data = tensorflow_datasets.load("glue/mrpc")
train_dataset = data["train"]
train_dataset = glue_convert_examples_to_features(train_dataset, tokenizer, 128, 'mrpc')
optimizer = tf.keras.optimizers.Adam(learning_rate=3e-5, epsilon=1e-08, clipnorm=1.0)
loss = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True)
metric = tf.keras.metrics.SparseCategoricalAccuracy('accuracy')
model.compile(optimizer=optimizer, loss=loss, metrics=[metric])
model.fit(train_dataset, epochs=3)
model = TFDistilBertForSequenceClassification.from_pretrained("distilbert-base-uncased")
tokenizer = DistilbertTokenizer.from_pretrained("distilbert-base-uncased")
2019 年 11 月 04 日 — 由 Hugging Face 团队 撰写
Hugging Face 是领先的 NLP 初创公司，超过一千家公司在生产中使用他们的库，包括必应、苹果和 Monzo。
本教程中使用的所有示例都可以在 Colab 上获取。相应的链接可在相应部分找到。
引言Hugging Face 是一家专注于 NLP 的初创公司，拥有庞大的开源社区，特别是在 Transformers 库方面。🤗/Transformers 是一个基于 Python 的库，它提供了一个 API 来使用许多知名的 Transformer 架构，例如 **BERT**、**RoBERTa**、**GPT-2** 或 **DistilBERT**，这些架构在各种 NLP 任务（如文本分类、信息提取、问答和文本生成）上获得了最先进的结果。这些架构预先训练了多组权重。使用 Transformers 入门只需要安装 pip 包
