Clothes Demand Prediction

• Predicting the demand of Clothes using LSTM and 3-layer neural network.
• To run the given codes, install Keras with tensorflow backend in your IPython shell (preferably Anaconda).

Business Problem

• Predicting the demand of various Clothing types in order to avoid inventory wastage.
• No input is at disposal, hence the input variables need to be forecasted and then the target variable is regressed through the forecasted input variable

Data Definition and Understanding

• Input variables

1. AvgSP - Average Selling Price of SKU
2. OP - Average Selling Price of various clothes
3. CustomerCount - Total GT Customers for the given SKU ( = CustomerCount + Missed Customers)

• Target Variable - ActualDemand of SKU ( = Ordered Quantity + Missed Demand)

• Summary Stats

Training and Test Datasets

• The last week of the complete dataset is considered for testing while the rest of the dataset is considered for training

Function to create Data Input to model

• AvgSP

1. @AvgSP is predicted using time series forecasting.
2. Long Short-Term Memory (Recurrent Neural Network) method is used for forecasting. The forecasting problem is now considered as a supervised learning problem where the input is the value prior to the target day.
3. LSTM is a special type of Neural Network which remembers information across long sequences to facilitate the forecasting.
4. Forecasting results

• CustomerCount

1. @CustomerCount is predicted using the same method as @AvgSP
2. Forecasting Results

Data Modelling

Model Name

• 3-layer Neural Network using Keras Library (tensorflow backend)
• The network is made up of 3 layers:

1. Input layer

• Takes input variables and converts them into input equation
• Parameters: no. of neurons (memory blocks) = 16, activation function = linear, weight initializer = normal distribution, kernel and activity regularizer = L1 (alpha = 0.1)

2. Hidden Layer

• The processing (optimization) takes place in this layer.
• Parameters: no. of neurons = 8, activation function = linear, weight initializer = normal distribution, kernel and activity regularizer = L1 (alpha = 0.1)

3. Output Layer

• Converts the processed results into a reverse scaled output.

Model Performance