-
原本一直使用LR模型作为CTR预估模型,最近正在调研LearningToRank模型(PaddlePaddle/models/ltr),随后便开始了对Paddle的学习。目前,遇到的最大问题是一轮训练(one pass)时间过长。
使用相同数据规模(7亿样本)、相同特征方案(离散化后的0/1稀疏特征3亿维)、相同cpu集群(40个节点),使用owlqn或LBFGS做优化的传统LR模型**# …
-
Hi guys,
I'm encountering an error. While I'm doing the task "performing feature extraction on resized images..." on a dataset of more than 55K images at almost 19K I'm facing out this:
`Traceb…
-
from gpytorch.inference import Inference
infer = Inference(prior_observation_model)
posterior_observation_model = infer.run(train_x, train_y, max_inference_steps=20)
*****************************…
-
Hi,
I am not sure how to create an intercept-only model after the recent changes to the package. Previously, this could be done with the arguments:
```
--feature-name-and-term-set-path='s3://b…
-
## 🐛 Bug
Torch fails to build (even with `BUILD_TEST=0`) on `/opt/pytorch/torch/csrc/Dtype.cpp`; looks like another instance of warning converted to errors:
```bash
[0mFirst, rewinding head to…
-
Compare to the original paper, my results using this implement is a little worse.
Is it caused by the Adam optimization rather than LBFGS in the original paper?
What is the difference of the the opt…
-
Example in examples/pes/neb exits with the following error:
```
Traceback (most recent call last):
File "/home/pjuda/bin/i-pi", line 101, in
main(args[0], options.do_yappi)
File "/home…
pjuda updated
6 years ago
-
ARMA model results have a convergence flag in `mod_res.mle_retvals`, but ARIMA model results do not have anything to indicate convergence. The flag is very useful during model selection, and in genera…
-
In the [tutorial of autoencoder](http://image.diku.dk/shark/sphinx_pages/build/html/rest_sources/tutorials/algorithms/autoencoders.html) the optimizer is `IRpropPlusFull` and I find that it's an "old"…
-
Continue to #702, the following steps are necessary to make GP module more accessible.
- [x] Support multidimensional output (currently, the output `y` is assumed to be 1D). This should not be comp…