I’m working on text classification problem. here is all the code import torch.nn as nn import torchtext.data as data import pandas as pd import torch import torch.nn.functional as F Text = data.Field(batch_first=True, include_lengths=True) LABEL = data.LabelField(dtype=torch.float, batch_first=Tr…

hidden has the shape [num_layers * num_directions, batch, hidden_size]. Since you are slicing it in dim0 and concatenating in dim1 here: x = torch.cat((hidden[-2, :, :], hidden[-1, :, :]), dim=1) the output shape of x will be [batch_size=32, 60], so you would need to increase the in_features of s…

RuntimeError: size mismatch, m1: [32 x 60], m2: [30 x 1] at ..\aten\src\TH/generic/THTensorMath.cpp:41

nlp

Bahaa_Kattan (Bahaa Kattan) October 9, 2020, 2:04pm 3

Thanks

Can you see this post, I got new error