decimer dataset

Figshare2024-03-20 更新2026-04-28 收录

化学图像识别

分子结构提取

数据链接：

https://figshare.com/articles/dataset/decimer_dataset/24591252 数据链接链接失效反馈

官方服务：

资源简介：

Dataset of hand drawn images of molecular depictions in tfrecord format together with instructions ready to train DECIMER.

创建时间：

2024-03-20

navanchauhan/decimer-data-mini

化学图像识别

文本到图像

PubChem 68K数据集包含68,996个分子的图像和对应的SMILES、SELFIES字符串。这些分子的分子量少于1500道尔顿，不包含反离子，仅由C, H, O, N, P, S, F, Cl, Br, I, Se和B元素组成，不包含氢的同位素，键的数量在3到40之间，不包含任何带电基团，仅包含隐式氢（功能基团除外），SMILES字符少于40个，且不允许立体化学。数据集来源于Decimer

Hugging Face2023-02-21 更新180

MOESM3 of ChemEngine: harvesting 3D chemical structures of supplementary data from PDF files

化学信息学

分子结构提取

Additional file 3. A comparative table consisting of interatomic bond distances computed via ChemEngine and the commercial software package. This material is available free of charge via the Internet

Figshare2017-02-21 更新60

MOESM4 of ChemEngine: harvesting 3D chemical structures of supplementary data from PDF files

化学信息学

分子结构提取

Additional file 4. Sample input file containing co-ordinates of molecules separated by comma delimiter.

NIAID Data Ecosystem40

Images dataset for Chemical Images Classifier

化学图像识别

化学图像分类

ChemIC: Images dataset for Chemical Images Classifier model

kaggle2024-08-29 更新80

jablonkagroup/MaCBench