📝 🌐 NLP From Scratch: Translation with a Sequence to Sequence Network and Attention 🌐

# For tips on running notebooks in Google Colab, see
# https://pytorch.org/tutorials/beginner/colab
%matplotlib inline

📝 🌐 NLP From Scratch: Translation with a Sequence to Sequence Network and Attention 🌐#

Author: Sean Robertson

This tutorials is part of a three-part series:

This is the third and final tutorial on doing NLP From Scratch, where we write our own classes and functions to preprocess the data to do our NLP modeling tasks.

In this project we will be teaching a neural network to translate from French to English.

… to varying degrees of success.

This is made possible by the simple but powerful idea of the sequence to sequence network, in which two recurrent neural networks work together to transform one sequence to another. An encoder network condenses an input sequence into a vector, and a decoder network unfolds that vector into a new sequence.

To improve upon this model we’ll use an attention mechanism, which lets the decoder learn to focus over a specific range of the input sequence.

Recommended Reading:

I assume you have at least installed PyTorch, know Python, and understand Tensors:

  • https://pytorch.org/ For installation instructions

  • /beginner/deep_learning_60min_blitz{.interpreted-text role=”doc”} to get started with PyTorch in general

  • /beginner/pytorch_with_examples{.interpreted-text role=”doc”} for a wide and deep overview

  • /beginner/former_torchies_tutorial{.interpreted-text role=”doc”} if you are former Lua Torch user

It would also be useful to know about Sequence to Sequence networks and how they work:

You will also find the previous tutorials on /intermediate/char_rnn_classification_tutorial{.interpreted-text role=”doc”} and /intermediate/char_rnn_generation_tutorial{.interpreted-text role=”doc”} helpful as those concepts are very similar to the Encoder and Decoder models, respectively.

Requirements

from __future__ import unicode_literals, print_function, division
from io import open
import unicodedata
import re
import random

import torch
import torch.nn as nn
from torch import optim
import torch.nn.functional as F

import numpy as np
from torch.utils.data import TensorDataset, DataLoader, RandomSampler

device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

Loading data files#

The data for this project is a set of many thousands of English to French translation pairs.

This question on Open Data Stack Exchange pointed me to the open translation site https://tatoeba.org/ which has downloads available at https://tatoeba.org/eng/downloads - and better yet, someone did the extra work of splitting language pairs into individual text files here: https://www.manythings.org/anki/

The English to French pairs are too big to include in the repository, so download to data/eng-fra.txt before continuing. The file is a tab separated list of translation pairs:

NOTE:

Download the data fromhereand extract it to the current directory.

!wget https://download.pytorch.org/tutorial/data.zip

--2025-03-11 22:01:36--  https://download.pytorch.org/tutorial/data.zip
Resolving download.pytorch.org (download.pytorch.org)... 108.139.29.36, 108.139.29.54, 108.139.29.73, ...
Connecting to download.pytorch.org (download.pytorch.org)|108.139.29.36|:443... connected.

HTTP request sent, awaiting response... 200 OK
Length: 2882130 (2.7M) [application/zip]
Saving to: ‘data.zip’


data.zip              0%[                    ]       0  --.-KB/s               

data.zip            100%[===================>]   2.75M  --.-KB/s    in 0.04s   

2025-03-11 22:01:36 (73.8 MB/s) - ‘data.zip’ saved [2882130/2882130]

!unzip /home/jca92/ENGR131_W25_dev/jupyterbook/week_10/lecture/data.zip -d data/

Archive:  /home/jca92/ENGR131_W25_dev/jupyterbook/week_10/lecture/data.zip
   creating: data/data/
  inflating: data/data/eng-fra.txt   

   creating: data/data/names/
  inflating: data/data/names/Arabic.txt  
  inflating: data/data/names/Chinese.txt  
  inflating: data/data/names/Czech.txt  
  inflating: data/data/names/Dutch.txt  
  inflating: data/data/names/English.txt  
  inflating: data/data/names/French.txt  
  inflating: data/data/names/German.txt  
  inflating: data/data/names/Greek.txt  
  inflating: data/data/names/Irish.txt  
  inflating: data/data/names/Italian.txt  
  inflating: data/data/names/Japanese.txt  
  inflating: data/data/names/Korean.txt  
  inflating: data/data/names/Polish.txt  
  inflating: data/data/names/Portuguese.txt  
  inflating: data/data/names/Russian.txt  
  inflating: data/data/names/Scottish.txt  
  inflating: data/data/names/Spanish.txt  
  inflating: data/data/names/Vietnamese.txt

Similar to the character encoding used in the character-level RNN tutorials, we will be representing each word in a language as a one-hot vector, or giant vector of zeros except for a single one (at the index of the word). Compared to the dozens of characters that might exist in a language, there are many many more words, so the encoding vector is much larger. We will however cheat a bit and trim the data to only use a few thousand words per language.

We’ll need a unique index per word to use as the inputs and targets of the networks later. To keep track of all this we will use a helper class called Lang which has word → index (word2index) and index → word (index2word) dictionaries, as well as a count of each word word2count which will be used to replace rare words later.

SOS_token = 0
EOS_token = 1


class Lang:
    def __init__(self, name):
        self.name = name
        self.word2index = {}
        self.word2count = {}
        self.index2word = {0: "SOS", 1: "EOS"}
        self.n_words = 2  # Count SOS and EOS

    def addSentence(self, sentence):
        for word in sentence.split(" "):
            self.addWord(word)

    def addWord(self, word):
        if word not in self.word2index:
            self.word2index[word] = self.n_words
            self.word2count[word] = 1
            self.index2word[self.n_words] = word
            self.n_words += 1
        else:
            self.word2count[word] += 1

The files are all in Unicode, to simplify we will turn Unicode characters to ASCII, make everything lowercase, and trim most punctuation.

# Turn a Unicode string to plain ASCII, thanks to
# https://stackoverflow.com/a/518232/2809427
def unicodeToAscii(s):
    return "".join(
        c for c in unicodedata.normalize("NFD", s) if unicodedata.category(c) != "Mn"
    )


# Lowercase, trim, and remove non-letter characters


def normalizeString(s):
    s = unicodeToAscii(s.lower().strip())
    s = re.sub(r"([.!?])", r" \1", s)
    s = re.sub(r"[^a-zA-Z!?]+", r" ", s)
    return s.strip()

To read the data file we will split the file into lines, and then split lines into pairs. The files are all English → Other Language, so if we want to translate from Other Language → English I added the reverse flag to reverse the pairs.

def readLangs(lang1, lang2, reverse=False):
    print("Reading lines...")

    # Read the file and split into lines
    lines = (
        open("data/data/%s-%s.txt" % (lang1, lang2), encoding="utf-8")
        .read()
        .strip()
        .split("\n")
    )

    # Split every line into pairs and normalize
    pairs = [[normalizeString(s) for s in l.split("\t")] for l in lines]

    # Reverse pairs, make Lang instances
    if reverse:
        pairs = [list(reversed(p)) for p in pairs]
        input_lang = Lang(lang2)
        output_lang = Lang(lang1)
    else:
        input_lang = Lang(lang1)
        output_lang = Lang(lang2)

    return input_lang, output_lang, pairs

Since there are a lot of example sentences and we want to train something quickly, we’ll trim the data set to only relatively short and simple sentences. Here the maximum length is 10 words (that includes ending punctuation) and we’re filtering to sentences that translate to the form “I am” or “He is” etc. (accounting for apostrophes replaced earlier).

MAX_LENGTH = 10

eng_prefixes = (
    "i am ",
    "i m ",
    "he is",
    "he s ",
    "she is",
    "she s ",
    "you are",
    "you re ",
    "we are",
    "we re ",
    "they are",
    "they re ",
)


def filterPair(p):
    return (
        len(p[0].split(" ")) < MAX_LENGTH
        and len(p[1].split(" ")) < MAX_LENGTH
        and p[1].startswith(eng_prefixes)
    )


def filterPairs(pairs):
    return [pair for pair in pairs if filterPair(pair)]

The full process for preparing the data is:

  • Read text file and split into lines, split lines into pairs

  • Normalize text, filter by length and content

  • Make word lists from sentences in pairs

def prepareData(lang1, lang2, reverse=False):
    input_lang, output_lang, pairs = readLangs(lang1, lang2, reverse)
    print("Read %s sentence pairs" % len(pairs))
    pairs = filterPairs(pairs)
    print("Trimmed to %s sentence pairs" % len(pairs))
    print("Counting words...")
    for pair in pairs:
        input_lang.addSentence(pair[0])
        output_lang.addSentence(pair[1])
    print("Counted words:")
    print(input_lang.name, input_lang.n_words)
    print(output_lang.name, output_lang.n_words)
    return input_lang, output_lang, pairs


input_lang, output_lang, pairs = prepareData("eng", "fra", True)
print(random.choice(pairs))

Reading lines...

Read 135842 sentence pairs
Trimmed to 11445 sentence pairs
Counting words...
Counted words:
fra 4601
eng 2991
['comme tu es mignonne', 'you are so cute']

The Seq2Seq Model#

A Recurrent Neural Network, or RNN, is a network that operates on a sequence and uses its own output as input for subsequent steps.

A Sequence to Sequence network, or seq2seq network, or Encoder Decoder network, is a model consisting of two RNNs called the encoder and decoder. The encoder reads an input sequence and outputs a single vector, and the decoder reads that vector to produce an output sequence.

Unlike sequence prediction with a single RNN, where every input corresponds to an output, the seq2seq model frees us from sequence length and order, which makes it ideal for translation between two languages.

Consider the sentence Je ne suis pas le chat noir → I am not the black cat. Most of the words in the input sentence have a direct translation in the output sentence, but are in slightly different orders, e.g. chat noir and black cat. Because of the ne/pas construction there is also one more word in the input sentence. It would be difficult to produce a correct translation directly from the sequence of input words.

With a seq2seq model the encoder creates a single vector which, in the ideal case, encodes the “meaning” of the input sequence into a single vector — a single point in some N dimensional space of sentences.

The Encoder#

The encoder of a seq2seq network is a RNN that outputs some value for every word from the input sentence. For every input word the encoder outputs a vector and a hidden state, and uses the hidden state for the next input word.

class EncoderRNN(nn.Module):
    def __init__(self, input_size, hidden_size, dropout_p=0.1):
        super(EncoderRNN, self).__init__()
        self.hidden_size = hidden_size

        self.embedding = nn.Embedding(input_size, hidden_size)
        self.gru = nn.GRU(hidden_size, hidden_size, batch_first=True)
        self.dropout = nn.Dropout(dropout_p)

    def forward(self, input):
        embedded = self.dropout(self.embedding(input))
        output, hidden = self.gru(embedded)
        return output, hidden

The Decoder#

The decoder is another RNN that takes the encoder output vector(s) and outputs a sequence of words to create the translation.

Simple Decoder#

In the simplest seq2seq decoder we use only last output of the encoder. This last output is sometimes called the context vector as it encodes context from the entire sequence. This context vector is used as the initial hidden state of the decoder.

At every step of decoding, the decoder is given an input token and hidden state. The initial input token is the start-of-string <SOS> token, and the first hidden state is the context vector (the encoder’s last hidden state).

class DecoderRNN(nn.Module):
    def __init__(self, hidden_size, output_size):
        super(DecoderRNN, self).__init__()
        self.embedding = nn.Embedding(output_size, hidden_size)
        self.gru = nn.GRU(hidden_size, hidden_size, batch_first=True)
        self.out = nn.Linear(hidden_size, output_size)

    def forward(self, encoder_outputs, encoder_hidden, target_tensor=None):
        batch_size = encoder_outputs.size(0)
        decoder_input = torch.empty(
            batch_size, 1, dtype=torch.long, device=device
        ).fill_(SOS_token)
        decoder_hidden = encoder_hidden
        decoder_outputs = []

        for i in range(MAX_LENGTH):
            decoder_output, decoder_hidden = self.forward_step(
                decoder_input, decoder_hidden
            )
            decoder_outputs.append(decoder_output)

            if target_tensor is not None:
                # Teacher forcing: Feed the target as the next input
                decoder_input = target_tensor[:, i].unsqueeze(1)  # Teacher forcing
            else:
                # Without teacher forcing: use its own predictions as the next input
                _, topi = decoder_output.topk(1)
                decoder_input = topi.squeeze(
                    -1
                ).detach()  # detach from history as input

        decoder_outputs = torch.cat(decoder_outputs, dim=1)
        decoder_outputs = F.log_softmax(decoder_outputs, dim=-1)
        return (
            decoder_outputs,
            decoder_hidden,
            None,
        )  # We return `None` for consistency in the training loop

    def forward_step(self, input, hidden):
        output = self.embedding(input)
        output = F.relu(output)
        output, hidden = self.gru(output, hidden)
        output = self.out(output)
        return output, hidden

I encourage you to train and observe the results of this model, but to save space we’ll be going straight for the gold and introducing the Attention Mechanism.

Attention Decoder#

If only the context vector is passed between the encoder and decoder, that single vector carries the burden of encoding the entire sentence.

Attention allows the decoder network to “focus” on a different part of the encoder’s outputs for every step of the decoder’s own outputs. First we calculate a set of attention weights. These will be multiplied by the encoder output vectors to create a weighted combination. The result (called attn_applied in the code) should contain information about that specific part of the input sequence, and thus help the decoder choose the right output words.

Calculating the attention weights is done with another feed-forward layer attn, using the decoder’s input and hidden state as inputs. Because there are sentences of all sizes in the training data, to actually create and train this layer we have to choose a maximum sentence length (input length, for encoder outputs) that it can apply to. Sentences of the maximum length will use all the attention weights, while shorter sentences will only use the first few.

Bahdanau attention, also known as additive attention, is a commonly used attention mechanism in sequence-to-sequence models, particularly in neural machine translation tasks. It was introduced by Bahdanau et al. in their paper titled Neural Machine Translation by Jointly Learning to Align and Translate. This attention mechanism employs a learned alignment model to compute attention scores between the encoder and decoder hidden states. It utilizes a feed-forward neural network to calculate alignment scores.

However, there are alternative attention mechanisms available, such as Luong attention, which computes attention scores by taking the dot product between the decoder hidden state and the encoder hidden states. It does not involve the non-linear transformation used in Bahdanau attention.

In this tutorial, we will be using Bahdanau attention. However, it would be a valuable exercise to explore modifying the attention mechanism to use Luong attention.

class BahdanauAttention(nn.Module):
    def __init__(self, hidden_size):
        super(BahdanauAttention, self).__init__()
        self.Wa = nn.Linear(hidden_size, hidden_size)
        self.Ua = nn.Linear(hidden_size, hidden_size)
        self.Va = nn.Linear(hidden_size, 1)

    def forward(self, query, keys):
        scores = self.Va(torch.tanh(self.Wa(query) + self.Ua(keys)))
        scores = scores.squeeze(2).unsqueeze(1)

        weights = F.softmax(scores, dim=-1)
        context = torch.bmm(weights, keys)

        return context, weights


class AttnDecoderRNN(nn.Module):
    def __init__(self, hidden_size, output_size, dropout_p=0.1):
        super(AttnDecoderRNN, self).__init__()
        self.embedding = nn.Embedding(output_size, hidden_size)
        self.attention = BahdanauAttention(hidden_size)
        self.gru = nn.GRU(2 * hidden_size, hidden_size, batch_first=True)
        self.out = nn.Linear(hidden_size, output_size)
        self.dropout = nn.Dropout(dropout_p)

    def forward(self, encoder_outputs, encoder_hidden, target_tensor=None):
        batch_size = encoder_outputs.size(0)
        decoder_input = torch.empty(
            batch_size, 1, dtype=torch.long, device=device
        ).fill_(SOS_token)
        decoder_hidden = encoder_hidden
        decoder_outputs = []
        attentions = []

        for i in range(MAX_LENGTH):
            decoder_output, decoder_hidden, attn_weights = self.forward_step(
                decoder_input, decoder_hidden, encoder_outputs
            )
            decoder_outputs.append(decoder_output)
            attentions.append(attn_weights)

            if target_tensor is not None:
                # Teacher forcing: Feed the target as the next input
                decoder_input = target_tensor[:, i].unsqueeze(1)  # Teacher forcing
            else:
                # Without teacher forcing: use its own predictions as the next input
                _, topi = decoder_output.topk(1)
                decoder_input = topi.squeeze(
                    -1
                ).detach()  # detach from history as input

        decoder_outputs = torch.cat(decoder_outputs, dim=1)
        decoder_outputs = F.log_softmax(decoder_outputs, dim=-1)
        attentions = torch.cat(attentions, dim=1)

        return decoder_outputs, decoder_hidden, attentions

    def forward_step(self, input, hidden, encoder_outputs):
        embedded = self.dropout(self.embedding(input))

        query = hidden.permute(1, 0, 2)
        context, attn_weights = self.attention(query, encoder_outputs)
        input_gru = torch.cat((embedded, context), dim=2)

        output, hidden = self.gru(input_gru, hidden)
        output = self.out(output)

        return output, hidden, attn_weights
NOTE:

There are other forms of attention that work around the lengthlimitation by using a relative position approach. Read about "localattention" in Effective Approaches to Attention-based Neural MachineTranslation.

Training#

Preparing Training Data#

To train, for each pair we will need an input tensor (indexes of the words in the input sentence) and target tensor (indexes of the words in the target sentence). While creating these vectors we will append the EOS token to both sequences.

def indexesFromSentence(lang, sentence):
    return [lang.word2index[word] for word in sentence.split(" ")]


def tensorFromSentence(lang, sentence):
    indexes = indexesFromSentence(lang, sentence)
    indexes.append(EOS_token)
    return torch.tensor(indexes, dtype=torch.long, device=device).view(1, -1)


def tensorsFromPair(pair):
    input_tensor = tensorFromSentence(input_lang, pair[0])
    target_tensor = tensorFromSentence(output_lang, pair[1])
    return (input_tensor, target_tensor)


def get_dataloader(batch_size):
    input_lang, output_lang, pairs = prepareData("eng", "fra", True)

    n = len(pairs)
    input_ids = np.zeros((n, MAX_LENGTH), dtype=np.int32)
    target_ids = np.zeros((n, MAX_LENGTH), dtype=np.int32)

    for idx, (inp, tgt) in enumerate(pairs):
        inp_ids = indexesFromSentence(input_lang, inp)
        tgt_ids = indexesFromSentence(output_lang, tgt)
        inp_ids.append(EOS_token)
        tgt_ids.append(EOS_token)
        input_ids[idx, : len(inp_ids)] = inp_ids
        target_ids[idx, : len(tgt_ids)] = tgt_ids

    train_data = TensorDataset(
        torch.LongTensor(input_ids).to(device), torch.LongTensor(target_ids).to(device)
    )

    train_sampler = RandomSampler(train_data)
    train_dataloader = DataLoader(
        train_data, sampler=train_sampler, batch_size=batch_size
    )
    return input_lang, output_lang, train_dataloader

Training the Model#

To train we run the input sentence through the encoder, and keep track of every output and the latest hidden state. Then the decoder is given the <SOS> token as its first input, and the last hidden state of the encoder as its first hidden state.

“Teacher forcing” is the concept of using the real target outputs as each next input, instead of using the decoder’s guess as the next input. Using teacher forcing causes it to converge faster but when the trained network is exploited, it may exhibit instability.

You can observe outputs of teacher-forced networks that read with coherent grammar but wander far from the correct translation -intuitively it has learned to represent the output grammar and can “pick up” the meaning once the teacher tells it the first few words, but it has not properly learned how to create the sentence from the translation in the first place.

Because of the freedom PyTorch’s autograd gives us, we can randomly choose to use teacher forcing or not with a simple if statement. Turn teacher_forcing_ratio up to use more of it.

def train_epoch(
    dataloader, encoder, decoder, encoder_optimizer, decoder_optimizer, criterion
):

    total_loss = 0
    for data in dataloader:
        input_tensor, target_tensor = data

        encoder_optimizer.zero_grad()
        decoder_optimizer.zero_grad()

        encoder_outputs, encoder_hidden = encoder(input_tensor)
        decoder_outputs, _, _ = decoder(encoder_outputs, encoder_hidden, target_tensor)

        loss = criterion(
            decoder_outputs.view(-1, decoder_outputs.size(-1)), target_tensor.view(-1)
        )
        loss.backward()

        encoder_optimizer.step()
        decoder_optimizer.step()

        total_loss += loss.item()

    return total_loss / len(dataloader)

This is a helper function to print time elapsed and estimated time remaining given the current time and progress %.

import time
import math


def asMinutes(s):
    m = math.floor(s / 60)
    s -= m * 60
    return "%dm %ds" % (m, s)


def timeSince(since, percent):
    now = time.time()
    s = now - since
    es = s / (percent)
    rs = es - s
    return "%s (- %s)" % (asMinutes(s), asMinutes(rs))

The whole training process looks like this:

  • Start a timer

  • Initialize optimizers and criterion

  • Create set of training pairs

  • Start empty losses array for plotting

Then we call train many times and occasionally print the progress (% of examples, time so far, estimated time) and average loss.

def train(
    train_dataloader,
    encoder,
    decoder,
    n_epochs,
    learning_rate=0.001,
    print_every=100,
    plot_every=100,
):
    start = time.time()
    plot_losses = []
    print_loss_total = 0  # Reset every print_every
    plot_loss_total = 0  # Reset every plot_every

    encoder_optimizer = optim.Adam(encoder.parameters(), lr=learning_rate)
    decoder_optimizer = optim.Adam(decoder.parameters(), lr=learning_rate)
    criterion = nn.NLLLoss()

    for epoch in range(1, n_epochs + 1):
        loss = train_epoch(
            train_dataloader,
            encoder,
            decoder,
            encoder_optimizer,
            decoder_optimizer,
            criterion,
        )
        print_loss_total += loss
        plot_loss_total += loss

        if epoch % print_every == 0:
            print_loss_avg = print_loss_total / print_every
            print_loss_total = 0
            print(
                "%s (%d %d%%) %.4f"
                % (
                    timeSince(start, epoch / n_epochs),
                    epoch,
                    epoch / n_epochs * 100,
                    print_loss_avg,
                )
            )

        if epoch % plot_every == 0:
            plot_loss_avg = plot_loss_total / plot_every
            plot_losses.append(plot_loss_avg)
            plot_loss_total = 0

    showPlot(plot_losses)

Plotting results#

Plotting is done with matplotlib, using the array of loss values plot_losses saved while training.

import matplotlib.pyplot as plt

plt.switch_backend("agg")
import matplotlib.ticker as ticker
import numpy as np


def showPlot(points):
    plt.figure()
    fig, ax = plt.subplots()
    # this locator puts ticks at regular intervals
    loc = ticker.MultipleLocator(base=0.2)
    ax.yaxis.set_major_locator(loc)
    plt.plot(points)

Evaluation#

Evaluation is mostly the same as training, but there are no targets so we simply feed the decoder’s predictions back to itself for each step. Every time it predicts a word we add it to the output string, and if it predicts the EOS token we stop there. We also store the decoder’s attention outputs for display later.

def evaluate(encoder, decoder, sentence, input_lang, output_lang):
    with torch.no_grad():
        input_tensor = tensorFromSentence(input_lang, sentence)

        encoder_outputs, encoder_hidden = encoder(input_tensor)
        decoder_outputs, decoder_hidden, decoder_attn = decoder(
            encoder_outputs, encoder_hidden
        )

        _, topi = decoder_outputs.topk(1)
        decoded_ids = topi.squeeze()

        decoded_words = []
        for idx in decoded_ids:
            if idx.item() == EOS_token:
                decoded_words.append("<EOS>")
                break
            decoded_words.append(output_lang.index2word[idx.item()])
    return decoded_words, decoder_attn

We can evaluate random sentences from the training set and print out the input, target, and output to make some subjective quality judgements:

def evaluateRandomly(encoder, decoder, n=10):
    for i in range(n):
        pair = random.choice(pairs)
        print(">", pair[0])
        print("=", pair[1])
        output_words, _ = evaluate(encoder, decoder, pair[0], input_lang, output_lang)
        output_sentence = " ".join(output_words)
        print("<", output_sentence)
        print("")

Training and Evaluating#

With all these helper functions in place (it looks like extra work, but it makes it easier to run multiple experiments) we can actually initialize a network and start training.

Remember that the input sentences were heavily filtered. For this small dataset we can use relatively small networks of 256 hidden nodes and a single GRU layer. After about 40 minutes on a MacBook CPU we’ll get some reasonable results.

NOTE:

If you run this notebook you can train, interrupt the kernel,evaluate, and continue training later. Comment out the lines where theencoder and decoder are initialized and run trainIters again.

hidden_size = 128
batch_size = 32

input_lang, output_lang, train_dataloader = get_dataloader(batch_size)

encoder = EncoderRNN(input_lang.n_words, hidden_size).to(device)
decoder = AttnDecoderRNN(hidden_size, output_lang.n_words).to(device)

train(train_dataloader, encoder, decoder, 80, print_every=5, plot_every=5)

Reading lines...
Read 135842 sentence pairs
Trimmed to 11445 sentence pairs
Counting words...
Counted words:
fra 4601
eng 2991
0m 32s (- 8m 2s) (5 6%) 1.5476
1m 1s (- 7m 12s) (10 12%) 0.6938
1m 31s (- 6m 37s) (15 18%) 0.3688
2m 1s (- 6m 4s) (20 25%) 0.2082
2m 31s (- 5m 32s) (25 31%) 0.1315
3m 1s (- 5m 1s) (30 37%) 0.0929
3m 30s (- 4m 31s) (35 43%) 0.0710
4m 0s (- 4m 0s) (40 50%) 0.0580
4m 30s (- 3m 30s) (45 56%) 0.0495
5m 0s (- 3m 0s) (50 62%) 0.0437
5m 30s (- 2m 30s) (55 68%) 0.0402
5m 59s (- 1m 59s) (60 75%) 0.0365
6m 29s (- 1m 29s) (65 81%) 0.0348
6m 59s (- 0m 59s) (70 87%) 0.0329
7m 29s (- 0m 29s) (75 93%) 0.0318
7m 58s (- 0m 0s) (80 100%) 0.0305

Set dropout layers to eval mode

encoder.eval()
decoder.eval()
evaluateRandomly(encoder, decoder)

> ce sont celles qui veulent s y rendre
= they are the ones who want to go
< they are the ones who want to go <EOS>

> il est tres sensible a la critique
= he is very sensitive to criticism
< he is very sensitive to criticism <EOS>

> j en suis presque sur
= i m almost sure of it
< i m almost sure of it wrong <EOS>

> vous etes tres talentueuse
= you re very talented
< you re very talented person <EOS>

> je veux presenter mes excuses
= i m willing to apologize
< i m willing to apologize <EOS>

> je suis content que ca vous rende heureux
= i m glad that makes you happy
< i m glad that makes you happy <EOS>

> elle va beaucoup mieux aujourd hui
= she s much better today
< she s much better today than yesterday <EOS>

> elle est accoutumee a veiller toute la nuit
= she is used to staying up all night
< she is used to staying up all night <EOS>

> vous etes vraiment geniaux
= you re really awesome
< you re really awesome <EOS>

> je suis certain
= i m sure
< i m sure of his success <EOS>

Visualizing Attention#

A useful property of the attention mechanism is its highly interpretable outputs. Because it is used to weight specific encoder outputs of the input sequence, we can imagine looking where the network is focused most at each time step.

You could simply run plt.matshow(attentions) to see attention output displayed as a matrix. For a better viewing experience we will do the extra work of adding axes and labels:

def showAttention(input_sentence, output_words, attentions):
    fig = plt.figure()
    ax = fig.add_subplot(111)
    cax = ax.matshow(attentions.cpu().numpy(), cmap="bone")
    fig.colorbar(cax)

    # Set up axes
    ax.set_xticklabels([""] + input_sentence.split(" ") + ["<EOS>"], rotation=90)
    ax.set_yticklabels([""] + output_words)

    # Show label at every tick
    ax.xaxis.set_major_locator(ticker.MultipleLocator(1))
    ax.yaxis.set_major_locator(ticker.MultipleLocator(1))

    plt.show()


def evaluateAndShowAttention(input_sentence):
    output_words, attentions = evaluate(
        encoder, decoder, input_sentence, input_lang, output_lang
    )
    print("input =", input_sentence)
    print("output =", " ".join(output_words))
    showAttention(input_sentence, output_words, attentions[0, : len(output_words), :])


evaluateAndShowAttention("il n est pas aussi grand que son pere")

evaluateAndShowAttention("je suis trop fatigue pour conduire")

evaluateAndShowAttention("je suis desole si c est une question idiote")

evaluateAndShowAttention("je suis reellement fiere de vous")

input = il n est pas aussi grand que son pere
output = he is not as tall as his father <EOS>
input = je suis trop fatigue pour conduire
output = i m too tired to drive <EOS>
input = je suis desole si c est une question idiote
output = i m sorry if this is a stupid question <EOS>

/tmp/ipykernel_630311/1690937169.py:8: UserWarning: set_ticklabels() should only be used with a fixed number of ticks, i.e. after set_ticks() or using a FixedLocator.
  ax.set_xticklabels([''] + input_sentence.split(' ') +
/tmp/ipykernel_630311/1690937169.py:10: UserWarning: set_ticklabels() should only be used with a fixed number of ticks, i.e. after set_ticks() or using a FixedLocator.
  ax.set_yticklabels([''] + output_words)
/tmp/ipykernel_630311/1690937169.py:8: UserWarning: set_ticklabels() should only be used with a fixed number of ticks, i.e. after set_ticks() or using a FixedLocator.
  ax.set_xticklabels([''] + input_sentence.split(' ') +
/tmp/ipykernel_630311/1690937169.py:10: UserWarning: set_ticklabels() should only be used with a fixed number of ticks, i.e. after set_ticks() or using a FixedLocator.
  ax.set_yticklabels([''] + output_words)
/tmp/ipykernel_630311/1690937169.py:8: UserWarning: set_ticklabels() should only be used with a fixed number of ticks, i.e. after set_ticks() or using a FixedLocator.
  ax.set_xticklabels([''] + input_sentence.split(' ') +
/tmp/ipykernel_630311/1690937169.py:10: UserWarning: set_ticklabels() should only be used with a fixed number of ticks, i.e. after set_ticks() or using a FixedLocator.
  ax.set_yticklabels([''] + output_words)
/tmp/ipykernel_630311/1690937169.py:8: UserWarning: set_ticklabels() should only be used with a fixed number of ticks, i.e. after set_ticks() or using a FixedLocator.
  ax.set_xticklabels([''] + input_sentence.split(' ') +
/tmp/ipykernel_630311/1690937169.py:10: UserWarning: set_ticklabels() should only be used with a fixed number of ticks, i.e. after set_ticks() or using a FixedLocator.
  ax.set_yticklabels([''] + output_words)

input = je suis reellement fiere de vous
output = i m really proud of you re <EOS>