@antinomy great! Sorry, there were many submissions on the last day.

We managed to reproduce this issue and we fixed it.

@captain-nidoran hi, all your strategies which will take part to the contest should be under the "In Contest" tab in the "Competition" section.

The migration "Candidates" -> "In Contest" was not immediate as we released minor improvements to the front-end side once the submission phase was over.

@magenta-grimer thank you @support 🙇 🙇 🙇 🙇 🙇 🙇
My compliments four your exceptional python skills

@sheikh Thanks, could you post the log you see on the notebook when the code fails? Thank you!

@support Thanks, mailing you!

@support Great! This route opens new possibilities in terms of model design. Thanks a lot!

@magenta-grimer sorry now it seems to work as it is....

Thank you for the report. The template has been updated.

@support Thank you once again. You guys are the best!

@magenta-grimer There are 2 things you might want to change:

1: the lookback_period is 365 but you want a 400-day SMA. This will only produce NaNs, so the boolean array sma20 < sma20_crypto will be False everywhere resulting in -1 weights. 2*365 as lookback does the trick for these settings.

2: Bitcoin is trading 24/7, futures aren't. Better use crypto.time.values instead of futures.time.values for the output of load_data.

There might be something else that I didn't catch but the resulting sharpe is at least close to what would be expected (1.109 with 5 and 385)

@support I have done that twice before my post, but now F_RY looks fine. There are several directories with scripts and notebooks I use with qnt, so maybe I deleted the wrong data-cache before...
Thanks for fixing the data!

@support Thank you for the prompt response. Time to get on with it. Thanks again.

@anshul96go Hello, when you submit, you have to select the competition type.

If you want to submit a Bitcoin system, you have to select "Cryptofutures".

If you want to submit a Futures system not including Bitcoin at all, you have to select "Futures".

@anshul96go Hi, sorry for the confusion:

the competition is still formally a "cryptofutures" one, you should get the data using qndata.cryptofutures.load_data(tail=period), as the timestamp are matched to the futures ones;

in connection to 1), data are daily level.

@support Yes, I can load now the stocks data. Thank you once again!

@antinomy Hi, sorry for the late answer. The problem is the size of files, we are working on a solution for avoiding this issue in the future.

@jeppe_and Hi, there are 4 options for backtesting:

processing all data in one pass. Very fast but you risk to overfit (by taking a global average for example); processing all data in many passes using the built-in backtester; same, but preserving the state at each iteration; for example, you can use the weights of yesterday to fix the weights today. In this case submission will take longer as parallelization is not possible; if yo use machine learning methods and you need to retrain your model periodically, you can use the ml extension for the call to the backtester.

@support Thank you!

Just out of curiousity I did some testing and it looks like the class actually was the culprit.
I submitted a simple strategy in 2 versions, one with a class and the other with a dictionary as state. The class version was rejected (exaclty like the one from my 1st post) and the dictionary version got accepted.