One-Liner Magic with Spark NLP: Deep Learning for Table Extraction 2/2

Sai Shailesh

We are going to use John Snow Labs’ NLP library with visual features installed to do whatever we did in the first part of the post Deep Learning based Table Extraction using Visual NLP 1/2, but just in a single line of code!

Imagine being able to predict with any of the 40,000+ pre-trained models or even train new ones, all in a single line of code! That’s the power of nlp.load().

All we have to do is simply load the model and predict on the image:

You can run the following code here: Colab

Using John Snow Labs one liner code

Now, to predict on the image:

Predictions on the image

Now, let’s look at the predictions generated above!

Warning::Spark Session already created, some configs may not take.
|    | col0                                | col1               | col2           |
|---:|:------------------------------------|:-------------------|:---------------|
|  0 |                                     | Number of.         |                |
|  1 |                                     | Series B           | Purchase       |
|  2 | Name of Shareholder                 | Preferred Shares _ | Amount         |
|  3 |                                     |                    | (US$)          |
|  4 | WuXi Healthcare Ventures            | 882,861            | 4,999,994.99   |
|  5 | 6 Dimensions Capital, L.P.          | 3,354,875          | 18,999,999.08  |
|  6 | 6 Dimensions Affiliates Fund, L.P.  | 176,572            | 999 997.87     |
|  7 | Graceful Beauty Limited             | 4,237,737          | 23,999,999.73  |
|  8 | Tetrad Ventures Pte Ltd             | 8,828,618          | 49,999 ,995.19 |
|  9 | Hikeo Biotech L.P.                  | 1,589,151          | 8,999,997.78   |
| 10 | Pure Progress International Limited | 1,765,723          | 9,999 995.64   |
| 11 | Kaitai International Funds SPC      | 882,861            | 4,999,994.99   |
| 12 | Taikang Kaitai (Cayman) Special     |                    |                |
| 13 | Opportunity I                       | 2,648,585          | 14,999,996.29  |
| 14 | CJS Medical Investment Limited      | 3,531,447          | 19,999,996.94  |
| 15 | SCC Growth IV Holdco G, Ltd.        | 5,297,171          | 29,999 998.25  |
| 16 | YF IV Checkpoint Limited            | 5,297,171          | 29,999 998.25  |
| 17 | HH CST Holdings Limited             | 1,765,723          | 9,999_,995.64  |
| 18 | ARCH Venture Fund IX, L.P.          | 441,430            | 2,499 994.67   |
| 19 | ARCH Venture Fund IX Overage, L.P.  | 1,324,292          | 7,499,995 .32  |
| 20 | Terra Magnum CST LLC                | 353,144            | 1,999,995.73   |
| 21 | 3W Partners Fund II, L.P.           | 882,861            | 4,999,994.99   |
| 22 | Huifu Investments Limited           | 882,861            | 4,999,994.99   |
| 23 | King Star Med LP                    | 1,765,723          | 9,999,995.64   |
| 24 | Total                               | 45,908,806         | 259,999,931.98 |

Indeed! With John Snow Labs NLP, table extraction becomes a breeze. Gone are the days of painstakingly creating a pipeline and making predictions. Now, we can accomplish all of this in a flash with just a single line of code! It’s not only efficient but also a real-time-saver.

Summary

In this article, we’ve taken a deep dive into the process of creating a pipeline for NER using John Snow Labs NLP with visual features installed. We’ve seen how the powerful tools provided by John Snow Labs can simplify this process, allowing us to perform complex tasks with just a single line of code. This not only makes the process more efficient but also saves us a significant amount of time.