Eduard Samokhvalov

Multiple models on one chart

4

### Expected Behavior Is there an elegant and simple way to visualize a few trading models plotted on 1 chart? In fact a lot of projects like mine are centered...

Problem: can't see ALL cores utilized while doing classification and predict catboost version: latest Operating System: Windows CPU: 32 core machine GPU: none Guys, I'm running this simple code from...

CatBoost to classify DataFrames

6

**Problem**: can I use CatBoost to classify around 40,000 different data series with various numeric properties (like stationarity, some other math metrics pre-calculated and assigned as features, some categorical labels,...

Additive vs Multiplicative mode to choose

1

Gents, I was reading a few articles in the internet on how to choose the best seasonality mode, but the answer is still not too clear to me. For example...

fb-prophet Python package installation on Windows 2016 failed with exit status 2

3

Gents, I have a typical issue installing fbprophet package on Windows with pip. Now it's something else :-) Below is the log, here're my specs: Windows Server 2019 64-bit OS,...

Does FB Prophet generated seasonality assume possible bias by historic periods that are outliers?

2

Hi gents. Imagine we have some really strong moves on the historic data (think BTC 2017 or market moves in October 2008), then 1) when we generate yearly seasonality with...

Problem: can a model stored into a file be updated with new data to retrain? catboost version: latest Operating System: any CPU: GPU:

not full CPU utilization

5

Problem: CPU load is very small catboost version: latest Operating System: windows server CPU: 32 cores GPU: no Guys, need an advise. I'm running the generation for 2 models, every...

simple signals from external file

1

Gents, I'm lookin for a way to use this format for my signals with vectorbt, it's EOD, so if predict >0 and predict>=threshold, then pnl=close-open, elif predict

https://youtu.be/usdEWSDisS0?t=3775 here Anna Veronica is saying that one can use own metric to evaluate model's performance. **Notebook** https://github.com/catboost/catboost/blob/671611e0a1ae47c03e7b2b52c06bdd91aeb23a5c/catboost/tutorials/events/2019_pydata_london/pydata_london_2019.ipynb#L4 **Section:** Metric evaluation on a new dataset Can you clarify what would...