Here's the task. You need to generate dates between two ranges, extract the year, extract the month and extract the quarter number. import pandas as pd start = '1/1/2018' end = '31/12/2020' df = pd.DataFrame(columns=['Year', 'Month', 'Quarter'] Le...
Attention mechanism is really fascinating. It tries to mimic human way of perceiving the world. In this post, I will provide visualizations on how attention mechanisms work on different simple problems. I will demonstrate them by reversing, right shi...
LSTM’s can learn from its input sequence when to use short term dependency and when to use long term dependency. In short term dependency it decides to clear the cell state. In long term dependency it decides not to clear the cell state. LSTM’s cont...
I use Self-Attention network for Sentiment Analysis. This model is equivalent to 1 head of Multi head Transformer Encoder. I use the IMDB dataset which contains 50K movie reviews which are marked positive or negative in sentiment. Architecture The i...
Extracting First Elements from Sublists. item1 = [['a', 1], ['b', 2], ['c', 3], ['d', 4]] list(map(operator.itemgetter(0), item1)) # ['a', 'b', 'c', 'd'] # Using Lambda list(map(lambda x: x[0], item1)) # ['a', 'b', 'c', 'd'] Extracting Last Eleme...
Categorical Cross Entropy and Sparse categorical Cross Entropy are loss functions used for classification tasks. The choice depends on the format of the label data. If the labels are one-hot encoded, categorical cross entropy is used. If the labels a...
How does one train ML or AI algorithms when dataset contains categorical variables? Algorithms only understand numerical values. That's why there are techniques known as Label Encoding or One-Hot Encoding. I find the concept cool. One dummy data for ...