nfelodcm 0.1.4__tar.gz → 0.1.7__tar.gz

{nfelodcm-0.1.4 → nfelodcm-0.1.7}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: nfelodcm
-Version: 0.1.4
+Version: 0.1.7
 Summary: Python package for loading and caching CSVs hosted on github into pandas dataframes
 Author-email: Robert Greer <nfl@robbygreer.com>
 Maintainer-email: Robert Greer <nfl@robbygreer.com>

{nfelodcm-0.1.4 → nfelodcm-0.1.7}/nfelodcm/Engine/Assignments/assignment.py

@@ -1,10 +1,24 @@
-from .assignments import game_id_repl
+from .assignments import (
+    game_id_repl, penalty_formatting, desc_based_plays
+)
 
 ## a dictionary that holds assignments and the columns they add ##
 assignments = {
     'game_id_repl' : {
         'func' : game_id_repl,
         'new_columns' : []
+    },
+    'penalty_formatting' : {
+        'func' : penalty_formatting,
+        'new_columns' : []
+    },
+    'desc_based_plays' : {
+        'func' : desc_based_plays,
+        'new_columns' : [
+            ('qb_dropback_all', 'float32'),
+            ('rush_attempt_all', 'float32'),
+            ('play_call', 'str'),
+        ]
     }
 }
 

nfelodcm-0.1.7/nfelodcm/Engine/Assignments/assignments/__init__.py

@@ -0,0 +1,3 @@
+from .fastr_team_repl import game_id_repl
+from .penalty_formatting import penalty_formatting
+from .desc_based_plays import desc_based_plays

nfelodcm-0.1.7/nfelodcm/Engine/Assignments/assignments/desc_based_plays.py

@@ -0,0 +1,57 @@
+import pandas as pd
+import numpy
+
+def desc_based_plays(df):
+    '''
+    Infers play types from description that may show up
+    differently in the standard columns, especially when 
+    there is a penalty
+    '''
+    ## infer pass play ##
+    df['desc_based_dropback'] = numpy.where(
+        (
+            (df['desc'].str.contains(' pass ', regex=False)) |
+            (df['desc'].str.contains(' sacked', regex=False)) |
+            (df['desc'].str.contains(' scramble', regex=False))
+        ),
+        1,
+        0
+    )
+    ## infer run ##
+    df['desc_based_run'] = numpy.where(
+        (
+            (~df['desc'].str.contains(' pass ', regex=False, na=False)) &
+            (~df['desc'].str.contains(' sacked', regex=False, na=False)) &
+            (~df['desc'].str.contains(' scramble', regex=False, na=False)) &
+            (~df['desc'].str.contains(' kicks ', regex=False, na=False)) &
+            (~df['desc'].str.contains(' punts ', regex=False, na=False)) &
+            (~df['desc'].str.contains(' field goal ', regex=False, na=False)) &
+            (df['desc'].str.contains(' to ', regex=False)) &
+            (df['desc'].str.contains(' for ', regex=False))
+        ),
+        1,
+        0
+    )
+    ## add new cols ##
+    df['qb_dropback_all'] = df[[
+        'qb_dropback', 'desc_based_dropback'
+    ]].max(axis=1)
+    df['rush_attempt_all'] = df[[
+        'rush_attempt', 'desc_based_run'
+    ]].max(axis=1)
+    ## create a specific play call field ##
+    df['play_call'] = numpy.where(
+        df['qb_dropback_all'] == 1,
+        'Pass',
+        numpy.where(
+            df['rush_attempt_all'] == 1,
+            'Run',
+            numpy.nan
+        )
+    )
+    ## clean up ##
+    df = df.drop(columns=[
+        'desc_based_dropback', 'desc_based_run'
+    ]).copy()
+    ## return ##
+    return df

{nfelodcm-0.1.4 → nfelodcm-0.1.7}/nfelodcm/Engine/Assignments/assignments/fastr_team_repl.py

@@ -14,6 +14,15 @@ def game_id_repl(df):
     """
     df['away_team'] = df['away_team'].replace(repl)
     df['home_team'] = df['home_team'].replace(repl)
+    ## add extra checks to make comparable with pbp ##
+    for col in [
+        'posteam', 'defteam', 'penalty_team',
+        'side_of_field', 'timeout_team', 'td_team',
+        'return_team'
+    ]:
+        if col in df.columns:
+            df[col] = df[col].replace(repl)
+    ## id col ##
     df['game_id'] = (
         df['season'].astype('str') + '_' +
         df['week'].astype('str').str.zfill(2) + '_' +

nfelodcm-0.1.7/nfelodcm/Engine/Assignments/assignments/penalty_formatting.py

@@ -0,0 +1,21 @@
+import pandas as pd
+import numpy
+
+def penalty_formatting(df):
+    """
+    Adds additional context to penalties
+    """
+    ## offensive and defensive penalties ##
+    df['off_penalty'] = numpy.where(
+        df['penalty_team'] == df['posteam'],
+        1,
+        0
+    )
+    df['def_penalty'] = numpy.where(
+        df['penalty_team'] == df['defteam'],
+        1,
+        0
+    )
+    ## remove nans from penalties to enable groupings ##
+    df['penalty_type'] = df['penalty_type'].fillna('No Penalty')
+    return df

{nfelodcm-0.1.4 → nfelodcm-0.1.7}/nfelodcm/Maps/pbp.json

@@ -16,7 +16,9 @@
     "type": "season",
     "start": 1999
   },
-  "assignments": [],
+  "assignments": [
+    "game_id_repl", "penalty_formatting", "desc_based_plays"
+  ],
   "map": {
     "play_id": "int32",
     "game_id": "object",

{nfelodcm-0.1.4 → nfelodcm-0.1.7}/nfelodcm/nfelodcm.py

@@ -1,5 +1,7 @@
 import pandas as pd
 import numpy
+import pathlib
+import json
 
 from .Engine import DCMTable
 import nfelodcm.nfelodcm.Utilities as utils
@@ -46,4 +48,23 @@ def get_map(url):
         print('"{0}": "{1}",'.format(
             col, dtype
         ))
-        
+
+def get_season_state(state_type='last_full_week'):
+    '''
+    Returns the season and week of the specified state type:
+       'last_full_week' (default): the last week in which no games are still unplayed
+       'last_partual_week': the last week where any game has been played
+       'next_week': the first week with no games played
+    '''
+    ## open global variables json ##
+    global_variables = None
+    with open('{0}/global_variables.json'.format(pathlib.Path(__file__).parent.resolve()), 'r') as fp:
+        ## load config ##
+        global_variables = json.load(fp)
+    ## check input ##
+    if state_type not in list(global_variables['season_states'].keys()):
+        raise Exception('UTILITY ERROR: {0} is not an available state_type. Available types: {1}'.format(
+            state_type, '\n     '.join(list(global_variables['season_states'].keys()))
+        ))
+    ## if state valid, return ##
+    return global_variables[state_type]['season'], global_variables[state_type]['week']

{nfelodcm-0.1.4 → nfelodcm-0.1.7}/nfelodcm.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: nfelodcm
-Version: 0.1.4
+Version: 0.1.7
 Summary: Python package for loading and caching CSVs hosted on github into pandas dataframes
 Author-email: Robert Greer <nfl@robbygreer.com>
 Maintainer-email: Robert Greer <nfl@robbygreer.com>

{nfelodcm-0.1.4 → nfelodcm-0.1.7}/nfelodcm.egg-info/SOURCES.txt

@@ -7,7 +7,9 @@ pyproject.toml
 ./nfelodcm/Engine/Assignments/__init__.py
 ./nfelodcm/Engine/Assignments/assignment.py
 ./nfelodcm/Engine/Assignments/assignments/__init__.py
+./nfelodcm/Engine/Assignments/assignments/desc_based_plays.py
 ./nfelodcm/Engine/Assignments/assignments/fastr_team_repl.py
+./nfelodcm/Engine/Assignments/assignments/penalty_formatting.py
 ./nfelodcm/Engine/Primatives/DataPull.py
 ./nfelodcm/Engine/Primatives/Freshness.py
 ./nfelodcm/Engine/Primatives/LocalIO.py

{nfelodcm-0.1.4 → nfelodcm-0.1.7}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "nfelodcm"
-version = "0.1.4"
+version = "0.1.7"
 readme = "README.md"
 requires-python = ">=3.11"
 license = {text = "MIT"}

nfelodcm-0.1.4/nfelodcm/Engine/Assignments/assignments/__init__.py

@@ -1 +0,0 @@
-from .fastr_team_repl import game_id_repl