FLEX to UTIL for NHL stacking on FD

Refactor stacking logic in app.py to conditionally apply calculations for 'Stack' and 'Size' columns based on their existence in the DataFrame. This change improves code robustness and prevents potential errors during data processing.

Add MMA stacking configuration and enhance error handling in position export logic in app.py. This update introduces new player positions for MMA and improves the robustness of the data processing by adding exception handling for sorting and deduplication operations.

recent changes

Update NHL stacking configurations in app.py to include detailed player positions for Draftkings and Fanduel. This change enhances lineup flexibility and aligns with current game strategies.

updating for recent changes

Refactor stacking logic in app.py to streamline team assignment calculations for player lineups. This update removes legacy stacking code and optimizes the application of team data, enhancing performance and maintainability.

Refactor portfolio filtering in app.py to convert entries to string and remove invalid values before saving to compressed format. This update enhances data integrity and ensures only valid entries are retained.

Remove debug print statements from app.py that were used for validating the working frame structure and content. This cleanup enhances code readability and maintains focus on core functionality.

Add debug prints for working frame in app.py to assist with data validation. This change helps verify the structure and content of the working frame after applying salary cap filtering.

Reintroduce salary cap filtering for working frame in app.py to ensure only eligible players are loaded. This update enhances data integrity and aligns with previous optimizations for memory efficiency.

Update portfolio management in app.py to reintroduce salary cap filtering for working frame. This change ensures that only eligible players are loaded into the working frame, enhancing data integrity and memory efficiency.

Refactor portfolio storage in app.py to filter by salary cap before saving to compressed format. This change optimizes memory usage and ensures only eligible entries are retained.

Update score cutoff in chunk_name_matching function and add PowerShell scripts for instance management

Added QB force option for micro

Enhance Lineup Edge calculation in predict_dupes.py by normalizing the edge values based on the maximum and minimum edges. This adjustment improves the scaling of the Lineup Edge, leading to more accurate portfolio predictions.

ping

Correct Lineup Edge calculation in predict_dupes.py by changing the adjustment from addition to subtraction based on duplicate ownership. This modification refines the edge computation, improving the accuracy of portfolio predictions.

Update Lineup Edge calculation in predict_dupes.py to include a normalized adjustment based on duplicate ownership. This change improves the accuracy of the edge computation by factoring in the average number of duplicates, enhancing overall portfolio predictions.

Refactor similarity score calculation in predict_dupes.py by removing the reference to duplicate adjustments. This change clarifies the computation of player selection diversity while preserving the overall prediction model's integrity.

Refactor Lineup Edge calculation in predict_dupes.py by removing the duplicate adjustment and simplifying the formula. This change enhances clarity in the computation while maintaining the integrity of the prediction model.

ping

Refactor Lineup Edge calculation in predict_dupes.py to incorporate both raw edge and adjusted edge values. This change enhances the accuracy of the prediction model by averaging the adjusted edge with the raw edge, improving overall portfolio predictions.

Adjust Lineup Edge_Raw calculation in predict_dupes.py to include a constant offset, enhancing the accuracy of the edge computation for portfolio predictions.

Enhance freq_format in app.py to include 'Lineup Edge_Raw', improving data representation for portfolio management.

Update excluded_cols in app.py to include 'Lineup Edge_Raw', enhancing the data handling for portfolio management.

Refactor Lineup Edge calculation in predict_dupes.py to utilize the raw edge value for duplicate adjustment. This change clarifies the computation and maintains the integrity of the prediction model.

Update Lineup Edge calculation in predict_dupes.py to separate raw edge computation. Adjusted the formula to remove the scaling based on duplicates, enhancing clarity and maintaining prediction accuracy.

ping

Refine Lineup Edge calculation in predict_dupes.py to adjust the edge based on the mean of duplicates. This update improves the accuracy of the prediction model by incorporating a scaling factor related to the average number of duplicates.

Refine Lineup Edge calculation in predict_dupes.py to average the adjusted edge based on the number of duplicates. This update enhances the prediction model's accuracy and performance.

diverting

Refine Lineup Edge calculation in predict_dupes.py to improve accuracy. Adjusted the formula to incorporate a divisor of (Dupes + 1) and modified the scaling factor, enhancing the prediction model's performance.

ping

Refine Lineup Edge calculation in predict_dupes.py to enhance accuracy. Updated the formula to adjust the edge based on the number of duplicates, improving the prediction model's performance.

Refactor Lineup Edge calculation in predict_dupes.py to improve accuracy. Updated the formula to average the adjusted edge based on the number of duplicates, enhancing the overall performance of the prediction model.

Refactor DataFrame optimization in app.py to enhance memory efficiency. Disable categorical conversion for specific columns to avoid issues with exposure_spread, while implementing smarter handling for numeric types. Introduce a new function to prepare DataFrames for exposure_spread, ensuring compatibility and improved performance during calculations.

Enhance DataFrame optimization and vectorized calculations in app.py. Refine category conversion logic to exclude specific player columns and improve memory efficiency. Introduce a safe mapping function to handle NaN values for salary, median, and ownership calculations, ensuring better performance across various scenarios.

Implement vectorized calculations for salary, median, and ownership in app.py to enhance performance and memory efficiency. Refactor reassess_edge and stratification_function to minimize DataFrame copies and improve memory management. Update filtering logic to use boolean masks for better efficiency.

Refactor portfolio storage in session state to use compressed format directly

Add memory optimization functions and implement chunked processing for name matching in app.py

Add error handling for position replacement in CSV loading process for Fanduel site. This ensures that any exceptions during the replacement of 'D' with 'DST' are gracefully handled without interrupting the flow.

Update requirements.txt to include scipy and ensure matplotlib is listed correctly

Add memory-efficient ranking calculation in predict_dupes.py

Update diversity score calculation to use chunked processing for improved performance in predict_dupes.py

Implement chunked processing for player similarity score calculation to enhance memory efficiency in predict_dupes.py

Refactor session state management by removing unnecessary copy operations for origin and working portfolios.

Add smart memory management to optimize session state handling and include psutil in requirements

Refine position replacement logic in CSV loading process to apply only when site_var is 'Fanduel'.

ping