Updates beyond the control of this repository may lead to the tutorial becoming outdated or non-functional. After reading the tutorial, run main.py.
To utilize the SDBSDataExtractor, it is essential to access both the tesseract download tutorial and the pytesseract usage tutorial. Skipping on using the class renders the engine download and package usage unnecessary.
To setup and execute the *.js file, is required to access the WebPlotDigitizer tutorial.
If you possess the SDBS numbers for the compounds of interest, utilizing this function may be unnecessary. You have the option to directly modify an existing CSV file or create a new one with the same name.
Upon executing this class, a message will be displayed in the console indicating that the system is now awaiting mouse clicks. Please follow the mouse click procedure as illustrated in the video provided below.
Click_tutorial.mp4
Following this, you'll need to verify the validity of the mouse clicks via the console:
Retry the clicks? (y/n)
To ensure the clicks were accurately placed, you can inspect the temporary files folder, which should contain two images: one displaying the SDBS numbers and the other showing the compound names.
The numbers and names will be sequentially saved to the existing CSV file. To conclude the number and names extraction process, provide confirmation through the console:
Continue capturing? (y/n)
Each iteration of the extracted data will have been securely saved.
This class will download the infrared (IR) spectra made in liquid film of the desired compounds.
Make sure that the compound has a valid IR spectra; in the absence of such data, no image will be downloaded.
You will have the choice to increase the number of chrome instances to download more images in a faster pace.
Following the download, the images will be formatted for compatibility with WebPlotDigitizer's code.
The reference project enables batch extraction and can be tailored through the WebPlotDigitizer application. Adjustments can be made by saving a new .json file.
All data gathered during the process will be accessible in the IR_spectral_data folder.
In case of any of the images on the img_data folder turn out like the one below, the task is to delete those images and then rerun the code.
This function uses the Whittaker smoothing and the airPLS algorithm of baseline correction from Z.-M. Zhang, S. Chen, and Y.-Z. Liang, 2010 to correct the collected data. The data is stored in a sqlite3 .db file for future use.