Guofu Li
Guofu Li
I'm also working on the re-construction of the stock-concept matrix from external resources. However, since there would be quite a lot guess works involved, I don't think my re-construction can...
I'm interested in this research direction too. I think, possibly, classifiers, or mulitple classifiers, may give better training signals than MSE regression learning objective. Moreover, if the classic TopKDropout strategy...
I don't think this is quite neccessary, as any of the Processor can be applied to either type of column.
In fact, `DropnaLabel` and `CSRankNorm` are usually used in the example implementations. I assume that, if you want to have some learning objectives other than Rank IC, you may also...
In most benchmark examples, `CSRankNorm` is used to create the `RankIC` learning label. But in other cases, you may also use this processor to make the feature column more robust...
According to my understanding, after reading the code, it uses some way to approximate the market capitalization based weights.
可以修改provider_uri来切换数据集，修改handler主要是从数据到因子和学习目标的加工过程
This error keeps poping up and is quite annoying.
> Try to change label to ["Ref($close, -1) / $close - 1"]. The original label is for Chinese market, which has different T0 trading rules. Also, US market is more...
I assume that the reason why feature values in the middle of the date range need to be nan-filled, is because the binary database requires the .bin file to be...