Page 220 - Ai Book - 10

Modelling

We can train the model according to the filtered datasets. Here, you can use a regression model in which the
dataset is fed as a dataframe and is trained accordingly. Let us understand with the help of an example. Suppose
you want to train a model for three days. As defined above, regression is a supervised Learning model which
works on continuous data. Here, we can divide the dataset into the ratio 20:10 which means 20 days are used
for training purpose and 10 days are used for testing purpose. The testing phase is also known as the Evaluation
phase in which we are able to check whether the model produce the desired outcome or not.

Evaluation

As you know, the evaluation phase is the last stage of the AI Project life Cycle. In this phase, we can easily
evaluate the outcome of the trained model by the following steps given below:

u Input Initial Data: The data should be fed into the trained model such as the name of the food items and
the quantity produced for the same.

u Input Additional Data: The additional data should also be fed into the trained model such as quantity of the
leftover food, number of absentees etc.

u Model Functioning: The trained model starts functioning on the basis of the data entries fed by you.
u Prediction: The Model predicts the quantity of food to be prepared for the next day.
u Comparison: In this step, the trained model starts prediction on the basis of testing dataset value.

u Testing: The model is tested for 10 testing datasets kept aside while training.
u Final Comparison: After testing, the trained model starts prediction values of the testing dataset as
compared to the actual values.

u Check Accuracy: If the prediction value matches with the actual values, the model is said to be accurate.

u Deployment: Once the model is able to produce optimum result, then it is ready to be deployed in the
restaurant for real-time usage.

TYPES OF DATA
You have learnt that the term ‘data’ is the collection of raw facts and figures. This data converts into processed
information using machines. Similarly, an AI model predicts optimum results on the basis of data which is being
fed by the programmer in different formats. Some of the commonly used data formats are as follows:

u Spreadsheet: In general, all data should be stored in tabular form. On the computer, we can mostly use a
spreadsheet application program such as MS Excel, Calc, etc. The term ‘Spreadsheet’ refers to a computer
program used for accounting and recording data using rows and columns into which information can be
entered.

u CSV: CSV is an acronym of Comma Separated Values which allows data to be saved in a tabular format.
CSVs look like a garden-variety spreadsheet but with a .csv extension.CSV files can be used with most any
spreadsheet program, such as Microsoft Excel or Google Spreadsheets. They differ from other spreadsheet
file types because you can only have a single sheet in a file, they can not save cells, columns, or rows.

u SQL: SQL is an acronym of Structured Query Language. It is a database computer language designed for the
retrieval and management of data in a relational database.

u ZIP: ZIP format is an archive file format. In Archive file format, you create a file that contains multiple files
along with metadata. An archive file format is used to collect multiple data files together into a single file.
This is done for simply compressing the files to use less storage space.

94
94