Text Classification with a Feedforward Network

develop a Feedforward neural network for text classification.

implement:

Text processing methods for transforming raw text data into input vectors for your network (2)

A Feedforward network consisting of:

One-hot input layer mapping words into an Embedding weight matrix (2 )
One hidden layer computing the mean embedding vector of all words in input followed by a ReLU activation
function (2 )
Output layer with a softmax activation. (2 )

The Stochastic Gradient Descent (SGD) algorithm with back-propagation to learn the weights of your Neural network.
Your algorithm should:
Use (and minimise) the Categorical Cross-entropy loss function (2 )
Perform a Forward pass to compute intermediate outputs (5 )
Perform a Backward pass to compute gradients and update all sets of weights (12 )
Implement and use Dropout after each hidden layer for regularisation (4 )

Discuss how did you choose hyperparameters? You can tune the learning rate (hint: choose small values), embedding
size {e.g. 50, 300, 500}, the dropout rate {e.g. 0.2, 0.5} and the learning rate. Please use tables or graphs to show training
and validation performance for each hyperparameter combination (5 ).

After training a model, plot the learning process (i.e. training and validation loss in each epoch) using a line plot and
report accuracy. Does your model overfit, underfit or is about right? (2 ).

Re-train your network by using pre-trained embeddings (GloVe) trained on large corpora. Instead of randomly initialising
the embedding weights matrix, you should initialise it with the pre-trained weights. During training, you should not
update them (i.e. weight freezing) and backprop should stop before computing gradients for updating embedding
weights. Report results by performing hyperparameter tuning and plotting the learning process. Do you get better
performance? (7 ).

Extend you Feedforward network by adding more hidden layers (e.g. one more or two). How does it affect the
performance? Note: You need to repeat hyperparameter tuning, but the number of combinations grows exponentially.
Therefore, you need to choose a subset of all possible combinations (8 )

Provide well documented and commented code describing all of your choices. In general, you are free to make decisions
about text processing (e.g. punctuation, numbers, vocabulary size) and hyperparameter values. Need justifications and
discussion for all of your choices (5 ).

Data
The data you will use for the task is a subset of the AG News Corpus and you can find it in the ./data_topic folder in CSV
format:

data_topic/train.csv : contains 2,400 news articles, 800 for each class to be used for training.
data_topic/dev.csv : contains 150 news articles, 50 for each class to be used for hyperparameter selection and
monitoring the training process.
data_topic/test.csv : contains 900 news articles, 300 for each class to be used for testing.

Pre-trained Embeddings
You can download pre-trained GloVe embeddings trained on Common Crawl (840B tokens, 2.2M vocab, cased, 300d vectors,
2.03 GB download) from here. No need to unzip, the file is large.

Save Memory
To save RAM, when you finish each experiment you can delete the weights of your network using del W followed by
Python's garbage collector gc.collect()

Instructions

1 of 10 23/04/2021, 05:04
 You are advised to follow the code structure given in this notebook by completing all given funtions. You can also write any
auxilliary/helper functions (and arguments for the functions) that you might need but note that you can provide a full
solution without any such functions. Similarly, you can just use only the packages imported below but you are free to use any
functionality from the Python Standard Library, NumPy, SciPy (excluding built-in softmax funtcions) and Pandas. You are not
allowed to use any third-party library such as Scikit-learn (apart from metric functions already provided), NLTK, Spacy,
Keras, Pytorch etc.. You should mention if you've used Windows to write and test your code because we mostly use Unix
based machines for marking (e.g. Ubuntu, MacOS).

There is no single correct answer on what your accuracy should be, but correct implementations usually achieve F1-scores
around 80\% or higher. The quality of the analysis of the results is as important as the accuracy itself.

import pandas as pd
import numpy as np
from collections import Counter
import re
import matplotlib.pyplot as plt
from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score
import random
from time import localtime, strftime
from scipy.stats import spearmanr,pearsonr
import zipfile
import gc

# fixing random seed for reproducibility

random.seed(123)
np.random.seed(123)

Transform Raw texts into training and development data

First, you need to load the training, development and test sets from their corresponding CSV files (tip: you can use Pandas
dataframes).

# dev = pd.read_csv("data_topic/dev.csv",header=0)
dev = pd.read_csv("data_topic/dev.csv", names=['label', 'text'])
train = pd.read_csv("data_topic/train.csv" ,header=0)
test = pd.read_csv("data_topic//test.csv")
dev.head()

label text

0 1 BAGHDAD, Iraq - An Islamic militant group that...

1 1 Parts of Los Angeles international airport are...

2 1 AFP - Facing a issue that once tripped up his ...

3 1 The leader of militant Lebanese group Hezbolla...

4 1 JAKARTA : ASEAN finance ministers ended a meet...

# raw_text = dev. iloc[:, 1]

# raw_text = dev['text']
raw_text=dev.text.tolist()
labels=dev.iloc[:,0].values

Create input representations

To train your Feedforward network, you first need to obtain input representations given a vocabulary. One-hot encoding
requires large memory capacity. Therefore, we will instead represent documents as lists of vocabulary indices (each word
corresponds to a vocabulary index).

Text Pre-Processing Pipeline

To obtain a vocabulary of words. You should:

tokenise all texts into a list of unigrams (tip: you can re-use the functions from Assignment 1)
remove stop words (using the one provided or one of your preference)

2 of 10 23/04/2021, 05:04
 remove unigrams appearing in less than K documents
use the remaining to create a vocabulary of the top-N most frequent unigrams in the entire corpus.
stop_words = ['a','in','on','at','and','or',
'to', 'the', 'of', 'an', 'by',
'as', 'is', 'was', 'were', 'been', 'be',
'are','for', 'this', 'that', 'these', 'those', 'you', 'i', 'if',
'it', 'he', 'she', 'we', 'they', 'will', 'have', 'has',
'do', 'did', 'can', 'could', 'who', 'which', 'what',
'but', 'not', 'there', 'no', 'does', 'not', 'so', 've', 'their',
'his', 'her', 'they', 'them', 'from', 'with', 'its']

Unigram extraction from a document

You first need to implement the extract_ngrams function. It takes as input:

x_raw : a string corresponding to the raw text of a document

ngram_range : a tuple of two integers denoting the type of ngrams you want to extract, e.g. (1,2) denotes extracting
unigrams and bigrams.
token_pattern : a string to be used within a regular expression to extract all tokens. Note that data is already
tokenised so you could opt for a simple white space tokenisation.
stop_words : a list of stop words
vocab : a given vocabulary. It should be used to extract specific features.

and returns:

a list of all extracted features.

def extract_ngrams(x_raw, ngram_range=(1,3), token_pattern=r'\b[A-Za-z][A-Za-z]+\b',

stop_words=[], vocab=set()):

tokenRE = re.compile(token_pattern)

# first extract all unigrams by tokenising

x_uni = [w for w in tokenRE.findall(str(x_raw).lower(),) if w not in stop_words]

# this is to store the ngrams to be returned

x = []

if ngram_range[0]==1:
x = x_uni

# generate n-grams from the available unigrams x_uni

ngrams = []
for n in range(ngram_range[0], ngram_range[1]+1):

# ignore unigrams
if n==1: continue

# pass a list of lists as an argument for zip

arg_list = [x_uni]+[x_uni[i:] for i in range(1, n)]

# extract tuples of n-grams using zip

# for bigram this should look: list(zip(x_uni, x_uni[1:]))
# align each item x[i] in x_uni with the next one x[i+1].
# Note that x_uni and x_uni[1:] have different lenghts
# but zip ignores redundant elements at the end of the second list
# Alternatively, this could be done with for loops
x_ngram = list(zip(*arg_list))
ngrams.append(x_ngram)

for n in ngrams:
for t in n:
x.append(t)

if len(vocab)>0:
x = [w for w in x if w in vocab]

return x

Create a vocabulary of n-grams

3 of 10 23/04/2021, 05:04
 Then the get_vocab function will be used to (1) create a vocabulary of ngrams; (2) count the document frequencies of
ngrams; (3) their raw frequency. It takes as input:

X_raw : a list of strings each corresponding to the raw text of a document

ngram_range : a tuple of two integers denoting the type of ngrams you want to extract, e.g. (1,2) denotes extracting
unigrams and bigrams.
token_pattern : a string to be used within a regular expression to extract all tokens. Note that data is already
tokenised so you could opt for a simple white space tokenisation.
stop_words : a list of stop words
min_df : keep ngrams with a minimum document frequency.
keep_topN : keep top-N more frequent ngrams.

and returns:

vocab : a set of the n-grams that will be used as features.

df : a Counter (or dict) that contains ngrams as keys and their corresponding document frequency as values.
ngram_counts : counts of each ngram in vocab

def get_vocab(X_raw, ngram_range=(1,3), token_pattern=r'\b[A-Za-z][A-Za-z]+\b',

min_df=0, keep_topN=0,
stop_words=[]):

tokenRE = re.compile(token_pattern)

df = Counter()
ngram_counts = Counter()
vocab = set()

# interate through each raw text

for x in X_raw:

x_ngram = extract_ngrams(x, ngram_range=ngram_range, token_pattern=token_pattern,

stop_words=stop_words)

#update doc and ngram frequencies

df.update(list(set(x_ngram)))
ngram_counts.update(x_ngram)

# obtain a vocabulary as a set.

# Keep elements with doc frequency > minimum doc freq (min_df)
# Note that df contains all te
vocab = set([w for w in df if df[w]>=min_df])

# keep the top N most freqent

if keep_topN>0:
vocab = set([w[0] for w in ngram_counts.most_common(keep_topN)
if w[0] in vocab])

return vocab, df, ngram_counts

Now you should use get_vocab to create your vocabulary and get document and raw frequencies of unigrams:

vocab, df, ngram_counts = get_vocab(raw_text)

print(type(vocab))

<class 'set'>

len(vocab)

9914

Then, you need to create vocabulary id -> word and word -> vocabulary id dictionaries for reference:

4 of 10 23/04/2021, 05:04
 dict1 = [*vocab]
dict_idword = {}
dict_wordid = {}

for i in range(len(dict1)):
dict_idword[i] = dict1[i]

for i in range(len(dict1)):
dict_wordid[dict1[i]] = i

Convert the list of unigrams into a list of vocabulary indices

Storing actual one-hot vectors into memory for all words in the entire data set is prohibitive. Instead, we will store word
indices in the vocabulary and look-up the weight matrix. This is equivalent of doing a dot product between an one-hot vector
and the weight matrix.

First, represent documents in train, dev and test sets as lists of words in the vocabulary:

raw_text
len(raw_text)

150

dev_words = []
for x in raw_text:
token_pattern1=r'\b[A-Za-z][A-Za-z]+\b'
tokenRE1 = re.compile(token_pattern1)
x_t = [w for w in tokenRE1.findall(str(x).lower(),) if w in vocab]
dev_words.append(x_t)

Then convert them into lists of indices in the vocabulary:

Put the labels Y for train, dev and test sets into arrays:

Network Architecture
Your network should pass each word index into its corresponding embedding by looking-up on the embedding matrix and
then compute the first hidden layer h1 :

1
h1 = ∑ Wie , i ∈ x
|x| i

where |x| is the number of words in the document and W e is an embedding matrix |V | × d, |V | is the size of the vocabulary
and d the embedding size.

Then h1 should be passed through a ReLU activation function:

a1 = relu(h1 )

Finally the hidden layer is passed to the output layer:

y = softmax(a1 W )

where W is a matrix d × |Y|, |Y| is the number of classes.

During training, a1 should be multiplied with a dropout mask vector (elementwise) for regularisation before it is passed to the
output layer.

You can extend to a deeper architecture by passing a hidden layer to another one:

hi = ai−1 Wi

ai = relu(hi )

5 of 10 23/04/2021, 05:04
 Network Training
First we need to define the parameters of our network by initiliasing the weight matrices. For that purpose, you should
implement the network_weights function that takes as input:

vocab_size : the size of the vocabulary

embedding_dim : the size of the word embeddings
hidden_dim : a list of the sizes of any subsequent hidden layers. Empty if there are no hidden layers between the
average embedding and the output layer
num_classes : the number of the classes for the output layer

and returns:

W : a dictionary mapping from layer index (e.g. 0 for the embedding matrix) to the corresponding weight matrix
initialised with small random numbers (hint: use numpy.random.uniform with from -0.1 to 0.1)

Make sure that the dimensionality of each weight matrix is compatible with the previous and next weight matrix, otherwise
you won't be able to perform forward and backward passes. Consider also using np.float32 precision to save memory.

def network_weights(vocab_size=1000, embedding_dim=300,

hidden_dim=[], num_classes=3, init_val = 0.5):

return W

W = network_weights(vocab_size=3,embedding_dim=4,hidden_dim=[2], num_classes=2)

Then you need to develop a softmax function (same as in Assignment 1) to be used in the output layer.

It takes as input z (array of real numbers) and returns sig (the softmax of z )

def softmax(z):

return sig

Now you need to implement the categorical cross entropy loss by slightly modifying the function from Assignment 1 to
depend only on the true label y and the class probabilities vector y_preds :

def categorical_loss(y, y_preds):

return l

Then, implement the relu function to introduce non-linearity after each hidden layer of your network (during the forward
pass):

relu(zi ) = max(zi , 0)

and the relu_derivative function to compute its derivative (used in the backward pass):

relu_derivative(zi )=0, if zi <=0, 1 otherwise.

Note that both functions take as input a vector z

Hint use .copy() to avoid in place changes in array z

def relu(z):

return a

def relu_derivative(z):

return dz

During training you should also apply a dropout mask element-wise after the activation function (i.e. vector of ones with a
random percentage set to zero). The dropout_mask function takes as input:

size : the size of the vector that we want to apply dropout

6 of 10 23/04/2021, 05:04
 dropout_rate : the percentage of elements that will be randomly set to zeros

and returns:

dropout_vec : a vector with binary values (0 or 1)

def dropout_mask(size, dropout_rate):

return dropout_vec

print(dropout_mask(10, 0.2))
print(dropout_mask(10, 0.2))

[1. 1. 0. 1. 1. 1. 1. 1. 0. 1.]
[1. 1. 1. 1. 0. 1. 1. 0. 1. 1.]

Now you need to implement the forward_pass function that passes the input x through the network up to the output
layer for computing the probability for each class using the weight matrices in W . The ReLU activation function should be
applied on each hidden layer.

x : a list of vocabulary indices each corresponding to a word in the document (input)

W : a list of weight matrices connecting each part of the network, e.g. for a network with a hidden and an output layer:
W[0] is the weight matrix that connects the input to the first hidden layer, W[1] is the weight matrix that connects the
hidden layer to the output layer.
dropout_rate : the dropout rate that is used to generate a random dropout mask vector applied after each hidden
layer for regularisation.

and returns:

out_vals : a dictionary of output values from each layer: h (the vector before the activation function), a (the resulting
vector after passing h from the activation function), its dropout mask vector; and the prediction vector (probability for
each class) from the output layer.

def forward_pass(x, W, dropout_rate=0.2):

out_vals = {}

h_vecs = []
a_vecs = []
dropout_vecs = []

return out_vals

The backward_pass function computes the gradients and updates the weights for each matrix in the network from the
output to the input. It takes as input

x : a list of vocabulary indices each corresponding to a word in the document (input)

y : the true label
W : a list of weight matrices connecting each part of the network, e.g. for a network with a hidden and an output layer:
W[0] is the weight matrix that connects the input to the first hidden layer, W[1] is the weight matrix that connects the
hidden layer to the output layer.
out_vals : a dictionary of output values from a forward pass.
learning_rate : the learning rate for updating the weights.
freeze_emb : boolean value indicating whether the embedding weights will be updated.

and returns:

W : the updated weights of the network.

Hint: the gradients on the output layer are similar to the multiclass logistic regression.

7 of 10 23/04/2021, 05:04
 def backward_pass(x, y, W, out_vals, lr=0.001, freeze_emb=False):

return W

Finally you need to modify SGD to support back-propagation by using the forward_pass and backward_pass functions.

The SGD function takes as input:

X_tr : array of training data (vectors)

Y_tr : labels of X_tr
W : the weights of the network (dictionary)
X_dev : array of development (i.e. validation) data (vectors)
Y_dev : labels of X_dev
lr : learning rate
dropout : regularisation strength
epochs : number of full passes over the training data
tolerance : stop training if the difference between the current and previous validation loss is smaller than a threshold
freeze_emb : boolean value indicating whether the embedding weights will be updated (to be used by the backward
pass function).
print_progress : flag for printing the training progress (train/validation loss)

and returns:

weights : the weights learned

training_loss_history : an array with the average losses of the whole training set after each epoch
validation_loss_history : an array with the average losses of the whole development set after each epoch

def SGD(X_tr, Y_tr, W, X_dev=[], Y_dev=[], lr=0.001,

dropout=0.2, epochs=5, tolerance=0.001, freeze_emb=False,
print_progress=True):

return W, training_loss_history, validation_loss_history

Now you are ready to train and evaluate your neural net. First, you need to define your network using the
network_weights function followed by SGD with backprop:

W = network_weights(vocab_size=len(vocab),embedding_dim=300,
hidden_dim=[], num_classes=3)

for i in range(len(W)):
print('Shape W'+str(i), W[i].shape)

W, loss_tr, dev_loss = SGD(X_tr, Y_tr,

W,
X_dev=X_dev,
Y_dev=Y_dev,
lr=0.001,
dropout=0.2,
freeze_emb=False,
tolerance=0.01,
epochs=100)

Plot the learning process:

Compute accuracy, precision, recall and F1-Score:

preds_te = [np.argmax(forward_pass(x, W, dropout_rate=0.0)['y'])

for x,y in zip(X_te,Y_te)]

print('Accuracy:', accuracy_score(Y_te,preds_te))
print('Precision:', precision_score(Y_te,preds_te,average='macro'))
print('Recall:', recall_score(Y_te,preds_te,average='macro'))
print('F1-Score:', f1_score(Y_te,preds_te,average='macro'))

Discuss how did you choose model hyperparameters ?

8 of 10 23/04/2021, 05:04
 Use Pre-trained Embeddings
Now re-train the network using GloVe pre-trained embeddings. You need to modify the backward_pass function above to
stop computing gradients and updating weights of the embedding matrix.

Use the function below to obtain the embedding martix for your vocabulary. Generally, that should work without any
problem. If you get errors, you can modify it.

def get_glove_embeddings(f_zip, f_txt, word2id, emb_size=300):

w_emb = np.zeros((len(word2id), emb_size))

with zipfile.ZipFile(f_zip) as z:
with z.open(f_txt) as f:
for line in f:
line = line.decode('utf-8')
word = line.split()[0]

if word in vocab:
emb = np.array(line.strip('\n').split()[1:]).astype(np.float32)
w_emb[word2id[word]] +=emb
return w_emb

w_glove = get_glove_embeddings("glove.840B.300d.zip","glove.840B.300d.txt",word2id)

First, initialise the weights of your network using the network_weights function. Second, replace the weigths of the
embedding matrix with w_glove . Finally, train the network by freezing the embedding weights:

preds_te = [np.argmax(forward_pass(x, W, dropout_rate=0.0)['y'])

for x,y in zip(X_te,Y_te)]

print('Accuracy:', accuracy_score(Y_te,preds_te))
print('Precision:', precision_score(Y_te,preds_te,average='macro'))
print('Recall:', recall_score(Y_te,preds_te,average='macro'))
print('F1-Score:', f1_score(Y_te,preds_te,average='macro'))

Discuss how did you choose model hyperparameters ?

Extend to support deeper architectures

Extend the network to support back-propagation for more hidden layers. You need to modify the backward_pass function
above to compute gradients and update the weights between intermediate hidden layers. Finally, train and evaluate a
network with a deeper architecture. Do deeper architectures increase performance?

preds_te = [np.argmax(forward_pass(x, W, dropout_rate=0.0)['y'])

for x,y in zip(X_te,Y_te)]

print('Accuracy:', accuracy_score(Y_te,preds_te))
print('Precision:', precision_score(Y_te,preds_te,average='macro'))
print('Recall:', recall_score(Y_te,preds_te,average='macro'))
print('F1-Score:', f1_score(Y_te,preds_te,average='macro'))

Discuss how did you choose model hyperparameters ?

Full Results
Add your final results here:

9 of 10 23/04/2021, 05:04
 Model Precision Recall F1-Score Accuracy

Average Embedding

Average Embedding (Pre-trained)

Average Embedding (Pre-trained) + X hidden layers

Please discuss why your best performing model is better than the rest.

10 of 10 23/04/2021, 05:04