seq2seq优化思路

发表于 |
字数统计: 208 | 阅读时长 ≈ 1

问题：
1. 程序添加测试集对模型进行离线验证，加上评价指标rmse
上次写的测试部分，数据来自训练集。
2. 添加网格搜索和交叉验证

RMSE函数

‘’’
from sklearn.metrics import mean_squared_error
import numpy as np
import sklearn.metrics as sm

def rmse(y_true, y_pred):
return np.sqrt(mean_squared_error(y_true, y_pred))

‘’’

需要搜索的参数：

1
2
3
4
5
6
7
8
9
10
batch_size = 5  # Low value used for live demo purposes - 100 and 1000 would be possible too, crank that up!
hidden_dim = 12  # Count of hidden neurons in the recurrent units.
layers_stacked_count = 2  # Number of stacked recurrent cells, on the neural depth axis.


learning_rate = 0.007  # Small lr helps not to diverge during training.
nb_iters = 300  # How many times we perform a training step (therefore how many times we show a batch).
lr_decay = 0.92  # default: 0.9 . Simulated annealing.
momentum = 0.5  # default: 0.0 . Momentum technique in weights update
lambda_l2_reg = 0.003  # L2 regularization of weights - avoids overfitting

交叉验证资料

k可不可以放到网格搜索里面呢？

https://zhuanlan.zhihu.com/p/24825503