Dynamic Theshold Model

[FireFlyTy]

• [x] Solution Validation and adjusting
• [x] Basis for results visualisation
• [x] Hyper Parameters turning for Expert Mode
• [x] Hyper Parameters turning for Simple Mode
• [x] Simple/Expert Mode wrapper
• [x] Model Save/Load
• [x] Manually Detected Periods
• [x] Results Interpretation

[FireFlyTy]

Some changes

1. Parallelisation of some operations
2. Time series preparation functions for model training and applying
3. Model Applying function
4. Adjusted function for thresholds aggregation
5. Correction and Scale coefficients were integrated to the model applying procedure, so that it's possible to adjust thresholds without model re-training
6. Parameters for "Simple mode" were proposed:

   High sensitivity

   agg_th = 0.5, local_trend = 0.1, k = 0.1, similar = 0.01

   Medium sensitivity

   agg_th = 0.6, local_trend = 0.3, k = 0.1, similar = 0.1

   Low sensitivity

   agg_th = 0.8, local_trend = 0.5, k = 0.1, similar = 0.3

[FireFlyTy]

1. Two-sides anomalies detection has been added
2. Model Training with Simple and Expert Modes support
3. Model Training and Applying (save + load)
4. Visualization based on time series as xts object