qiong-yang/fastei

Ultra-fast and Accurate Spectrum Matching

fastei's Issues

test.py不能体现出论文中36.7的召回率

您好，项目中的test.py似乎并不能体现出论文中Top1 36.7%的召回率以及Top10 80.4%的召回率，data/query spectra/文件夹也不存在

spec2vec training code

Hi Yang,

Will you add the code you use to train the spec2vec model? I would like to improve in spectrum embedding. Thanks!

example code logic / error

I've testing the code and basically I have this output of the query molecules but the answer is quite strange :

here the smiles in the test folder

here the answer based on the indexes found :

and If I look at only the first molecule (ie I[0]):

I don't see any accurate matching at all, can you tell me what can be the issue ?
here the matching index I have [2117858, 83971, 1358966, 1944009, 46292, 213015, 901693, 253884, 173348, 1360939]

matching_index=[I[0][0],I[1][0],I[2][0],I[3][0],I[4][0],I[5][1],I[6][8],I[7][14],I[8][17],I[9][99]]
print(matching_index)

ModuleNotFoundError: No module named 'data_process'

Hello, may I ask how this module was obtained

Batch import and export

Dear developers,

thank you very much for this really nice software. I started using the GUI and have few questions:

1. Is it possible to import directly several compounds at once? Maybe in MGF format? (for now I am importing every compounds one by one with 1 CSV files per compound)
2. Is it possible to export a result summary for all the imported compounds?
3. I was wondering if in the spectra plot we could see the mass of the fragments? I have high-resolution data and would like to assess directly how good a fragment match is (are the predicted fragments using high-resolution data?)

Cheers
Jean