--- title: TAGNN++ keywords: fastai sidebar: home_sidebar summary: "Mitheran et. al. Improved Representation Learning for Session-based Recommendation. arXiv, 2021." description: "Mitheran et. al. Improved Representation Learning for Session-based Recommendation. arXiv, 2021." nb_path: "nbs/models/tagnn_pp.ipynb" ---
{% raw %}
{% endraw %} {% raw %}
{% endraw %} {% raw %}

class Attention_GNN[source]

Attention_GNN(hidden_size, step=1) :: Module

Base class for all neural network modules.

Your models should also subclass this class.

Modules can also contain other Modules, allowing to nest them in a tree structure. You can assign the submodules as regular attributes::

import torch.nn as nn
import torch.nn.functional as F

class Model(nn.Module):
    def __init__(self):
        super(Model, self).__init__()
        self.conv1 = nn.Conv2d(1, 20, 5)
        self.conv2 = nn.Conv2d(20, 20, 5)

    def forward(self, x):
        x = F.relu(self.conv1(x))
        return F.relu(self.conv2(x))

Submodules assigned in this way will be registered, and will have their parameters converted too when you call :meth:to, etc.

:ivar training: Boolean represents whether this module is in training or evaluation mode. :vartype training: bool

{% endraw %} {% raw %}
{% endraw %} {% raw %}

class TAGNN_PP[source]

TAGNN_PP(opt) :: Module

Base class for all neural network modules.

Your models should also subclass this class.

Modules can also contain other Modules, allowing to nest them in a tree structure. You can assign the submodules as regular attributes::

import torch.nn as nn
import torch.nn.functional as F

class Model(nn.Module):
    def __init__(self):
        super(Model, self).__init__()
        self.conv1 = nn.Conv2d(1, 20, 5)
        self.conv2 = nn.Conv2d(20, 20, 5)

    def forward(self, x):
        x = F.relu(self.conv1(x))
        return F.relu(self.conv2(x))

Submodules assigned in this way will be registered, and will have their parameters converted too when you call :meth:to, etc.

:ivar training: Boolean represents whether this module is in training or evaluation mode. :vartype training: bool

{% endraw %} {% raw %}
{% endraw %}

Training a session-based recommender using TAGNN++

{% raw %}
class Args():
    dataset = 'sample'
    batchSize = 100 # input batch size
    hiddenSize = 100 # hidden state size
    epoch = 5 # the number of epochs to train for
    lr = 0.001 # learning rate')  # [0.001, 0.0005, 0.000
    lr_dc = 0.1 # learning rate decay rate
    lr_dc_step = 3 # the number of steps after which the learning rate decay
    l2 = 1e-5 # l2 penalty')  # [0.001, 0.0005, 0.0001, 0.00005, 0.0000
    step = 1 # gnn propogation steps
    patience = 10 # the number of epoch to wait before early stop 
    nonhybrid = True # only use the global preference to predict
    validation = True # validation
    valid_portion = 0.1 # split the portion of training set as validation set
    n_node = 310

args = Args()
{% endraw %} {% raw %}
def to_cuda(input_variable):
    if torch.cuda.is_available():
        return input_variable.cuda()
    else:
        return input_variable


def to_cpu(input_variable):
    if torch.cuda.is_available():
        return input_variable.cpu()
    else:
        return input_variable


def forward(model, i, data):
    alias_inputs, A, items, mask, targets = data.get_slice(i)
    alias_inputs = to_cuda(torch.Tensor(alias_inputs).long())
    items = to_cuda(torch.Tensor(items).long())
    A = to_cuda(torch.Tensor(A).float())
    mask = to_cuda(torch.Tensor(mask).long())
    hidden = model(items, A)

    def get(i): return hidden[i][alias_inputs[i]]
    seq_hidden = torch.stack([get(i)
                             for i in torch.arange(len(alias_inputs)).long()])

    return targets, model.compute_scores(seq_hidden, mask)


def train_test(model, train_data, test_data):
    model.scheduler.step()
    print('Start training: ', datetime.datetime.now())
    model.train()
    total_loss = 0.0
    slices = train_data.generate_batch(model.batch_size)

    from tqdm.notebook import tqdm
    for i, j in tqdm(zip(slices, np.arange(len(slices))), total=len(slices)):
        model.optimizer.zero_grad()
        targets, scores = forward(model, i, train_data)
        targets = to_cuda(torch.Tensor(targets).long())
        loss = model.loss_function(scores, targets - 1)
        loss.backward()
        model.optimizer.step()
        total_loss += loss.item()

        if j % int(len(slices) / 5 + 1) == 0:
            print('[%d/%d] Loss: %.4f' % (j, len(slices), loss.item()))

    print('\tLoss Value:\t%.3f' % total_loss)
    print('Start Prediction: ', datetime.datetime.now())

    model.eval()
    hit, mrr = [], []
    slices = test_data.generate_batch(model.batch_size)

    for i in slices:
        targets, scores = forward(model, i, test_data)
        sub_scores = scores.topk(20)[1]
        sub_scores = to_cpu(sub_scores).detach().numpy()

        for score, target, mask in zip(sub_scores, targets, test_data.mask):
            hit.append(np.isin(target - 1, score))
            if len(np.where(score == target - 1)[0]) == 0:
                mrr.append(0)
            else:
                mrr.append(1 / (np.where(score == target - 1)[0][0] + 1))

    hit = np.mean(hit) * 100
    mrr = np.mean(mrr) * 100
    return hit, mrr


def get_pos(seq_len):
    return torch.arange(seq_len).unsqueeze(0)


def str2bool(v):
    return v.lower() in ('true')


def split_validation(train_set, valid_portion):
    train_set_x, train_set_y = train_set
    n_samples = len(train_set_x)
    sidx = np.arange(n_samples, dtype='int32')
    np.random.shuffle(sidx)
    n_train = int(np.round(n_samples * (1. - valid_portion)))
    valid_set_x = [train_set_x[s] for s in sidx[n_train:]]
    valid_set_y = [train_set_y[s] for s in sidx[n_train:]]
    train_set_x = [train_set_x[s] for s in sidx[:n_train]]
    train_set_y = [train_set_y[s] for s in sidx[:n_train]]

    return (train_set_x, train_set_y), (valid_set_x, valid_set_y)
{% endraw %} {% raw %}
import os
import pickle
import time
from torch.utils.tensorboard import SummaryWriter

from recohut.datasets.session import SampleSessionDataset, GraphData

import warnings
warnings.filterwarnings('ignore')

model_save_dir = 'saved/'
log_dir='saved/logs'
writer = SummaryWriter(log_dir=log_dir)
os.makedirs(log_dir, exist_ok=True)

_ = SampleSessionDataset('./session_ds')
train_data = pickle.load(open('./session_ds/processed/train.txt', 'rb'))

if args.validation:
    train_data, valid_data = split_validation(train_data, args.valid_portion)
    test_data = valid_data
else:
    test_data = pickle.load(open('./session_ds/processed/test.txt', 'rb'))

train_data = GraphData(train_data, shuffle=True)
test_data = GraphData(test_data, shuffle=False)

model = to_cuda(TAGNN_PP(args))

start = time.time()
best_result = [0, 0]
best_epoch = [0, 0]
bad_counter = 0

for epoch in range(args.epoch):
    print('-' * 50)
    print('Epoch: ', epoch)
    hit, mrr = train_test(model, train_data, test_data)
    flag = 0

    # Logging
    writer.add_scalar('epoch/recall', hit, epoch)
    writer.add_scalar('epoch/mrr', mrr, epoch)

    flag = 0

    if hit >= best_result[0]:
        best_result[0] = hit
        best_epoch[0] = epoch
        flag = 1
        torch.save(model, model_save_dir + 'epoch_' +
                    str(epoch) + '_recall_' + str(hit) + '_.pt')
    if mrr >= best_result[1]:
        best_result[1] = mrr
        best_epoch[1] = epoch
        flag = 1
        torch.save(model, model_save_dir + 'epoch_' +
                    str(epoch) + '_mrr_' + str(mrr) + '_.pt')

    print('Best Result:')
    print('\tRecall@20:\t%.4f\tMRR@20:\t%.4f\tEpoch:\t%d,\t%d' %
            (best_result[0], best_result[1], best_epoch[0], best_epoch[1]))

    bad_counter += 1 - flag

    if bad_counter >= args.patience:
        break

print('-' * 50)
end = time.time()
print("Running time: %f seconds" % (end - start))
--------------------------------------------------
Epoch:  0
Start training:  2021-12-23 11:37:30.078500
[0/11] Loss: 5.7335
[3/11] Loss: 5.7263
[6/11] Loss: 5.7291
[9/11] Loss: 5.7135
	Loss Value:	62.968
Start Prediction:  2021-12-23 11:37:31.807639
Best Result:
	Recall@20:	17.3554	MRR@20:	5.2499	Epoch:	0,	0
--------------------------------------------------
Epoch:  1
Start training:  2021-12-23 11:37:31.935884
[0/11] Loss: 5.7028
[3/11] Loss: 5.6784
[6/11] Loss: 5.6666
[9/11] Loss: 5.6629
	Loss Value:	62.498
Start Prediction:  2021-12-23 11:37:33.677782
Best Result:
	Recall@20:	19.8347	MRR@20:	5.2499	Epoch:	1,	0
--------------------------------------------------
Epoch:  2
Start training:  2021-12-23 11:37:33.802865
[0/11] Loss: 5.6609
[3/11] Loss: 5.6498
[6/11] Loss: 5.6462
[9/11] Loss: 5.6439
	Loss Value:	62.090
Start Prediction:  2021-12-23 11:37:35.502085
Best Result:
	Recall@20:	20.6612	MRR@20:	5.2499	Epoch:	2,	0
--------------------------------------------------
Epoch:  3
Start training:  2021-12-23 11:37:35.604826
[0/11] Loss: 5.6461
[3/11] Loss: 5.6324
[6/11] Loss: 5.6430
[9/11] Loss: 5.6417
	Loss Value:	61.997
Start Prediction:  2021-12-23 11:37:37.213415
Best Result:
	Recall@20:	21.4876	MRR@20:	5.2499	Epoch:	3,	0
--------------------------------------------------
Epoch:  4
Start training:  2021-12-23 11:37:37.333273
[0/11] Loss: 5.6551
[3/11] Loss: 5.6529
[6/11] Loss: 5.5994
[9/11] Loss: 5.6067
	Loss Value:	61.898
Start Prediction:  2021-12-23 11:37:38.855207
Best Result:
	Recall@20:	21.4876	MRR@20:	5.2499	Epoch:	4,	0
--------------------------------------------------
Running time: 8.897622 seconds
{% endraw %}