bupianlizhugui

Very slow in dimensionality reduction

1

I found a problem when I was practicing the code. When the amount of data was more than 1000, the speed of dimensionality reduction was very slow and I could...

Given a table name and a problem, how do models make predictions? I see the source code only gives the validation script

when I run： def _debug(self, model, sliced_data, output): for i, item in enumerate(tqdm.tqdm(sliced_data)): (_, history), = **model.compute_loss([item], debug=True)** output.write( json.dumps({ 'index': i, 'history': history, }) + '\n') output.flush() def compute_loss(self,...

您好作者，请问这部分逻辑能稍微解释下吗？type_index_in_token_ids这个变量有点迷，

1

https://github.com/qiufengyuyi/event_extraction/blob/f7daafce811ee9c865353d1a91a827427d7c3e6a/models/bert_event_type_classification.py#L99  type_index_in_token_ids这个变量是存储了原始句子语料+标签索引后的len长度，每加一个标签长度都做存储。为啥和batch_ids有关系呢？

这行代码是不是写错啦？token_ids = [token_ids_org[0]] + prefix_token_ids + [token_ids_org[1]] 这里，结尾后面应该是token_ids_org[-1]

1

https://github.com/qiufengyuyi/event_extraction/blob/f7daafce811ee9c865353d1a91a827427d7c3e6a/data_processing/event_prepare_data.py#L564

请问刘老师有可导入到neo4j图书库到点边数据格式文件吗？

3

请问刘老师有可导入到neo4j图书库到点边数据格式文件吗？ 另一方面 ，关于电商届的数据 是否有相关资源的推荐？ 非常感谢开源

from typing import List from PIL import Image import torch import numpy as np from uform import get_model, Modality import torch.nn as nn def get_image_embedding(images: List[Image.Image]): preprocessed = processor_image(images) embedding...

from uform import get_model, Modality import torch.nn as nn encoders, processors = get_model('unum-cloud/uform-vl-multilingual-v2',backend='torch') model_text = models[Modality.TEXT_ENCODER] model_image = models[Modality.IMAGE_ENCODER] processor_text = processors[Modality.TEXT_ENCODER] processor_image = processors[Modality.IMAGE_ENCODER] model_text.return_features = False model_image.return_features =...