4 votes

Solve tough clustering problem with overlapping clusters

The Problem is that many clustering algorithms focus on distances (between points, clusters and so on). Especially at the connection between the two desired clusters, distances between points are ...
Broele's user avatar
  • 1,362
4 votes

Pandas Profiling Not Working

Something to note is that the package name will soon change to ydata-profiling, so we should use the new name. This is the announcement on their Pypi site: ⚠️ ...
noe's user avatar
  • 26.6k
2 votes

Using boxplots of previous day’s percentage change in S&P index to predict today's return based on yesterday's movement

If the median for the 'up' bucket is much lower than for the 'down' bucket, this means that days that have a negative market return are more likely to be followed by a day with a positive return. This ...
Oxbowerce's user avatar
  • 7,507
2 votes

Affine 2D mapping in python

Bias term $c$ can be eliminated (eg by subtracting the averages from $A$ and $B$) and the approximation be framed as: $$B = M \cdot A$$ Using minimum sum of squared residual errors criterion (similar ...
Nikos M.'s user avatar
  • 2,333
2 votes
Accepted

what qualifies as a data leakage?

Data leakage occurs in cases when you train a model with data that is not available for future testing/inference; or when you use same piece of data for training, and then for validation and/or ...
Stefan Popov's user avatar
2 votes

what qualifies as a data leakage?

One definition of data leakage is providing the model with data during training that would not be available at a future prediction time. The variable "total target achieved/units purchased as on ...
Brian Spiering's user avatar
1 vote
Accepted

How should I plot 2D data as 3D plot?

What your boss means by 'split x-axis' is not clear for me either. Might be best to ask that person. Will your datapoints be like $\left(X,\,Y,\,Z\right)=\left(10, \,0,\,20\right)$, $\left(20,\,10,\,...
Cryo's user avatar
  • 533
1 vote
Accepted

ValueError: X has 54 features, but DecisionTreeClassifier is expecting 53 features as input

Where you do pred_set.drop(["Winner"], axis=1) you need to add inPlace=True, or assign to a new dataframe, otherwise nothing happens. ...
brewmaster321's user avatar
1 vote

Bibliographic references on statistical analysis of classification results in machine learning

Benavoli, Alessio, et al. "Time for a change: a tutorial for comparing multiple classifiers through Bayesian analysis." The Journal of Machine Learning Research 18.1 (2017): 2653-2688. This ...
Dave's user avatar
  • 3,843
1 vote

Is it essential to remove NULL values when doing EDA?

Like so many things, it’s going to depend on what you want to do with the data once it’s been cleaned. Your options include dropping the entire row where null values are present, dropping the columns ...
brewmaster321's user avatar
1 vote
Accepted

Analysing 1GB of 200K of data rows having 50 columns

There are multiple options if you want to extract insights from your data, some of which are: Python or R : If you are comfortable in either of the 2 languages, you can easily generate visualizations ...
spectre's user avatar
  • 2,065

Only top scored, non community-wiki answers of a minimum length are eligible