Sellopale/deepmindcodecontestssharegpt

Selloworkspace数据集是一个多任务数据集，涵盖文本分类、表格问答、问答、零样本分类、翻译、摘要、特征提取、文本生成、填充掩码、句子相似性、文本转语音、文本转音频、自动语音识别、音频转音频、音频分类、语音活动检测、深度估计、图像分类、目标检测、图像分割、文本转图像和图像转文本等多种任务。数据集主要应用于金融、法律和代码领域。数据集包含对话形式的conversations，其中包含from和value两个字段。训练集包含13134个样本，总大小为45256223字节。

The Selloworkspace dataset is a multi-task dataset covering various tasks such as text classification, table-question-answering, question-answering, zero-shot-classification, translation, summarization, feature-extraction, text-generation, fill-mask, sentence-similarity, text-to-speech, text-to-audio, automatic-speech-recognition, audio-to-audio, audio-classification, voice-activity-detection, depth-estimation, image-classification, object-detection, image-segmentation, text-to-image, and image-to-text. The dataset is primarily applied in the fields of finance, legal, and code. It features conversations in a dialog format, including from and value fields. The training set contains 13134 examples with a total size of 45256223 bytes.