distance_matrix(series1, series2, max_freq, norm) Generate distance matrix between two series. Parameters: series1 (float): Source series 1. series2 (float): Source series 2. max_freq (int): Maximum frequency to inpect or the size of the generated matrix. norm (string): Norm of the distance metric, default=`euclidean`, options=`euclidean`, `manhattan`, `max`. Returns: Matrix with distance values.
method normalize_distance(M) Normalizes a matrix within its Min-Max range. Namespace types: matrix<float> Parameters: M (matrix<float>): Source matrix. Returns: Normalized matrix.
method threshold(M, threshold) Updates the matrix with the condition `M(i,j) > threshold ? 1 : 0`. Namespace types: matrix<float> Parameters: M (matrix<float>): Source matrix. threshold (float) Returns: Cross matrix.
rolling_window(a, b, sample_size) An experimental alternative method to plot a recurrence_plot. Parameters: a (array<float>): Array with data. b (array<float>): Array with data. sample_size (int) Returns: Recurrence_plot matrix.
Hi Ricardo, Thank you so much!!! Quick Quest, what time frame should we look for, I got a study error - Error On Bar 4300 too many candles in history (6944). Thanks again!
RicardoSantos
⋅
@Ranedisk, the algorithm is extremely demanding so use on charts with reduced candles to avoid that
slowcoconut
⋅
I notice earlier you mention taking time to work with LLM... and I see mention of '~starling7b ' ... have you fine-tuned some LLM? If so, that's amazing.
RicardoSantos
⋅
@slowcoconut, nah.. just toying with it really, noting professional, starling is a solid model for quick QA with good outputs for its small size, you can look it up on huggingface
Thank you so much!!! Quick Quest, what time frame should we look for, I got a study error - Error On Bar 4300 too many candles in history (6944).
Thanks again!