def eval_rank(model, test_loader, lossfn, parallel, top_k):
model.eval()
HR, NDCG = [], []
for batch_id, batch in enumerate(test_loader):
u_idxs = batch[0].long().cuda()
i_idxs = batch[1].long().cuda()
predictions = model(u_idxs, i_idxs)
if parallel and torch.cuda.device_count() >1:
i_idxs = i_idxs.view(torch.cuda.device_count(), -1)
print(predictions)
for device_idx, prediction in enumerate(predictions):
device = torch.device('cuda:{}'.format(device_idx))
i_idx = i_idxs[device_idx, :].to(device)
_, indices = torch.topk(prediction, top_k)
recommends = torch.take(i_idx, indices).cpu().numpy().tolist()
gt_item = i_idx[0].item()
HR.append(hit(gt_item, recommends))
NDCG.append(ndcg(gt_item, recommends))
else:
_, indices = torch.topk(predictions, top_k)
recommends = torch.take(i_idxs, indices).cpu().numpy().tolist()
gt_item = i_idxs[0].item()
HR.append(hit(gt_item, recommends))
NDCG.append(ndcg(gt_item, recommends))
if batch_id % 240 == 0 :
print("-----------The timeStamp of evaluating batch {:03d}/{}".format(batch_id, len(test_loader)) + " is: " + time.strftime("%H: %M: %S", time.gmtime(time.time())))
return np.mean(HR), np.mean(NDCG)
def report_pos_neg(x):
# print(x)
# pos_itemIdx, recommends = x
# recommends = list(recommends)
# return hit(pos_itemIdx, recommends), ndcg(pos_itemIdx, recommends)
pos_itemIdx, itemIdxs, indices = x
pos_itemIdx = int(pos_itemIdx)
recommends = torch.take(itemIdxs, indices)
recommends = list(recommends)
return hit(pos_itemIdx, recommends), ndcg(pos_itemIdx, recommends)