TorchScript Model inference slow in python

[driazati]

The first pass does some extra work (details in the link below), if you warm up ScriptModel by calling it a few times first the time should improve.

Speed of Custom RNN is SUPER SLOW

Hi, Based on code here https://github.com/pytorch/pytorch/blob/master/benchmarks/fastrnns/custom_lstms.py I write an example to compare the cumputation capability of native lstm and custom lstm. But I found that the speed of custom lstm is 100 times slower than native lstm class. here is my test code: import torch.nn as nn import time from models.custom_lstm import LSTMLayer, LSTMCell, script_lstm, LSTMState input_size = 1024 cell_size =2048 batch_size =20 seq_len = 200 native_lstm=n…