Trjtdtk

What are the advantages and disadvantages of running one shots compared to campaigns?

I’m planning on buying a laser printer but concerned about the life cycle of toner in the machine

Is a vector space a subspace of itself?

Why do UK politicians seemingly ignore opinion polls on Brexit?

Short story: alien planet where slow students are executed

Prime joint compound before latex paint?

Are white and non-white police officers equally likely to kill black suspects?

Can I legally use front facing blue light in the UK?

Information to fellow intern about hiring?

How would photo IDs work for shapeshifters?

How to move the player while also allowing forces to affect it

What causes the sudden spool-up sound from an F-16 when enabling afterburner?

How to manage monthly salary

extract characters between two commas?

Email Account under attack (really) - anything I can do?

Is it wise to hold on to stock that has plummeted and then stabilized?

A poker game description that does not feel gimmicky

What do the Banks children have against barley water?

Does a dangling wire really electrocute me if I'm standing in water?

What is the command to reset a PC without deleting any files

"listening to me about as much as you're listening to this pole here"

Patience, young "Padovan"

How did the USSR manage to innovate in an environment characterized by government censorship and high bureaucracy?

How to make payment on the internet without leaving a money trail?

Predicting when features are time-dependent

1

$begingroup$

How to predict data with time-dependent features?

For example, I have to predict the result of a Mortal Combat game:

X(i) = [player1_id, player2_id, hero_of_player1, hero_of_player2]

Y(i) = 1 if player1 wins or 0 if player2 wins

I have a dataset, containing games from large period of time. And, of course, performance of each player can variate during that time. Second, there were a lot of game patches, and they tuned some heroes' abilities. As a result overall hero strength can also variate through time as well as certain hero-vs-hero match-ups.

How to track that changes considering also overall and time-dependent impact of player+hero, hero-vs-hero, player-vs-player features?

So far I'm using simple LSTM network on whole game history. So my data has a single sample and total number of steps equal to game history length:

X = data.reshape(1,len(data),len(data[0]) #len(data[0]) corresponds to 4 features for a single record in dataset, but it is onehot-encoded.

m = Sequential()
m.add(LSTM(25, input_shape=(None, len(data[0])), return_sequences=True)
m.add(TimeDistirbuted(Dense(1))

I've got about 60% performance on the real problem. And I think it might be better and I can tune layer sizes, regularize, add stacking LSTMs. But did I choose an adequate NN-structure? What are the most suitable NN-structures for that kind of problems? Sliding-windows and Convolutional LSTMs sounds promising, but I don't have enough intuition about them.

deep-learning keras time-series lstm

share|improve this question

edited Mar 29 at 13:11

Glorfindel

1511210

asked Mar 29 at 7:27

farynaafarynaa

62

$endgroup$

add a comment | 

1

$begingroup$

How to predict data with time-dependent features?

For example, I have to predict the result of a Mortal Combat game:

X(i) = [player1_id, player2_id, hero_of_player1, hero_of_player2]

Y(i) = 1 if player1 wins or 0 if player2 wins

I have a dataset, containing games from large period of time. And, of course, performance of each player can variate during that time. Second, there were a lot of game patches, and they tuned some heroes' abilities. As a result overall hero strength can also variate through time as well as certain hero-vs-hero match-ups.

How to track that changes considering also overall and time-dependent impact of player+hero, hero-vs-hero, player-vs-player features?

So far I'm using simple LSTM network on whole game history. So my data has a single sample and total number of steps equal to game history length:

X = data.reshape(1,len(data),len(data[0]) #len(data[0]) corresponds to 4 features for a single record in dataset, but it is onehot-encoded.

m = Sequential()
m.add(LSTM(25, input_shape=(None, len(data[0])), return_sequences=True)
m.add(TimeDistirbuted(Dense(1))

I've got about 60% performance on the real problem. And I think it might be better and I can tune layer sizes, regularize, add stacking LSTMs. But did I choose an adequate NN-structure? What are the most suitable NN-structures for that kind of problems? Sliding-windows and Convolutional LSTMs sounds promising, but I don't have enough intuition about them.

deep-learning keras time-series lstm

share|improve this question

edited Mar 29 at 13:11

Glorfindel

1511210

asked Mar 29 at 7:27

farynaafarynaa

62

$endgroup$

add a comment | 

$begingroup$

How to predict data with time-dependent features?

For example, I have to predict the result of a Mortal Combat game:

X(i) = [player1_id, player2_id, hero_of_player1, hero_of_player2]

Y(i) = 1 if player1 wins or 0 if player2 wins

I have a dataset, containing games from large period of time. And, of course, performance of each player can variate during that time. Second, there were a lot of game patches, and they tuned some heroes' abilities. As a result overall hero strength can also variate through time as well as certain hero-vs-hero match-ups.

How to track that changes considering also overall and time-dependent impact of player+hero, hero-vs-hero, player-vs-player features?

So far I'm using simple LSTM network on whole game history. So my data has a single sample and total number of steps equal to game history length:

X = data.reshape(1,len(data),len(data[0]) #len(data[0]) corresponds to 4 features for a single record in dataset, but it is onehot-encoded.

m = Sequential()
m.add(LSTM(25, input_shape=(None, len(data[0])), return_sequences=True)
m.add(TimeDistirbuted(Dense(1))

I've got about 60% performance on the real problem. And I think it might be better and I can tune layer sizes, regularize, add stacking LSTMs. But did I choose an adequate NN-structure? What are the most suitable NN-structures for that kind of problems? Sliding-windows and Convolutional LSTMs sounds promising, but I don't have enough intuition about them.

deep-learning keras time-series lstm

share|improve this question

edited Mar 29 at 13:11

Glorfindel

1511210

asked Mar 29 at 7:27

farynaafarynaa

62

$endgroup$

X(i) = [player1_id, player2_id, hero_of_player1, hero_of_player2]

Y(i) = 1 if player1 wins or 0 if player2 wins

X = data.reshape(1,len(data),len(data[0]) #len(data[0]) corresponds to 4 features for a single record in dataset, but it is onehot-encoded.

m = Sequential()
m.add(LSTM(25, input_shape=(None, len(data[0])), return_sequences=True)
m.add(TimeDistirbuted(Dense(1))

share|improve this question

edited Mar 29 at 13:11

Glorfindel

1511210

asked Mar 29 at 7:27

farynaafarynaa

62

share|improve this question

edited Mar 29 at 13:11

Glorfindel

1511210

asked Mar 29 at 7:27

farynaafarynaa

62

edited Mar 29 at 13:11

Glorfindel

1511210

edited Mar 29 at 13:11

Glorfindel

1511210

asked Mar 29 at 7:27

farynaafarynaa

62

asked Mar 29 at 7:27

farynaafarynaa

62