-
I have the data of financial transactions and I am trying to fit a neural network to it. Here is some background on the data.
I have stored it in a dataframe which has date as index and transaction…
-
### Describe the bug
When looking at the representations of TransE before and after training multiple times using the training loop with `continue_training=True`, only the entity representation cha…
-
See https://github.com/atkinssamuel/cryptonet-research/blob/master/BaselineTiming/Models/Encrypted/encrypted_test.py
The summarize the nature of this repo:
- The problem being addressed was a regr…
-
optimizer中优化器支持哪几种？只有默认的Momentum优化器是正常的。将.yml配置文件中的优化器改成Adam会出现训练正常但评估过程“卡住”，不报错就一直卡在评估过程中。
-
Hi, I am trying to reproduce the training results I generated using google/seq2seq on tensorflow/nmt.
I noticed that standard hyperparams provided here lead to much higher BLEU score (15.9 vs. 21.…
ghost updated
6 years ago
-
Using Adam optimizer, not L2L for the CIFAR problem: If I print the cost after each epoch, it doesn't decrease over time running with learning rate .001, num_steps 100 num_epochs 100. However, printin…
-
Hi,
I made changes to your network. Added the batchnorm and xavier initializations, but i noticed you have used Adam optimizer while paper used SGD with decay and momentum, even the parameters are no…
-
### 请提出你的问题 Please ask your question
您好，在论文复现的过程中已经实现模型的正向对齐和损失函数对齐，反向从第二轮损失开始对不齐，通过打印torch和paddle的反向梯度，通过ReprodLogger进行diff验证，发现是使用的paddle.vision.resnet的预训练模型中的downsample层开始diff到了e-4级别，在这下面的梯度均为e-…
-
I am trying to use a following little complicated LSTM language model for text generation.
```python
# -*- coding: utf-8 -*-
import json
import os
os.environ["CUDA_VISIBLE_DEVICES"] = "0"
impo…
-
CPM 使用微调脚本训练， 不开启 --use-delta 这一选项，则出现如下错误：
Traceback (most recent call last):
File "finetune_cpm_bee.py", line 503, in
main()
File "finetune_cpm_bee.py", line 499, in main
finetune(…