baseballDataScience/trainPitcherModel.py at master · parkererickson/baseballDataScience · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
from pybaseball import statcast
import pandas as pd
import numpy as np
from pybaseball import statcast_pitcher
from pybaseball import playerid_lookup

from sklearn.preprocessing import LabelEncoder
from sklearn.model_selection import train_test_split
from imblearn.over_sampling import SMOTE
from keras.utils import to_categorical

from tensorflow import feature_column
from tensorflow.keras import layers
import tensorflow as tf

pid = playerid_lookup('berrios','jose')["key_mlbam"][0]
print(pid)
# get all available data
data = statcast_pitcher('2017-03-01', '2019-10-10', player_id = pid)

data = data[["pitch_type", "bat_score", "fld_score", "on_3b", "on_2b", "on_1b", "outs_when_up", "inning", "inning_topbot", "pitch_number", "p_throws", "balls", "strikes", "stand", "batter", "release_speed", "description"]]

data = data[data.pitch_type != 'EP']
data = data[data.pitch_type != 'PO']

data[["on_3b", "on_2b", "on_1b"]] = data[["on_3b", "on_2b", "on_1b"]].replace(np.nan, 0)
data.loc[data.on_3b > 0, "on_3b"] = 1
data.loc[data.on_2b > 0, "on_2b"] = 1
data.loc[data.on_1b > 0, "on_1b"] = 1
data = data.dropna()

data["prev_pitch"] = data["pitch_type"].shift(-1)
data["prev_batter"] = data["batter"].shift(-1)
data["release_speed"] = data["release_speed"].shift(-1)
data["description"] = data["description"].shift(-1)

def filterAB(row):
    if row["batter"] == row["prev_batter"]:
        return row["prev_pitch"]
    else:
        return "NB"

def filterSpeed(row):
    if row["batter"] == row["prev_batter"]:
        return row["release_speed"]
    else:
        return 0

def filterDesc(row):
    if row["batter"] == row["prev_batter"]:
        return row["description"]
    else:
        return "NB"

data["prev_pitch"] = data.apply(lambda row: filterAB(row), axis = 1)
data["release_speed"] = data.apply(lambda row: filterSpeed(row), axis=1)
data["description"] = data.apply(lambda row: filterDesc(row), axis=1)

data = data.drop("prev_batter", 1)
data = data.drop("batter", 1)

data = pd.concat([data,pd.get_dummies(data["prev_pitch"], prefix='prev_pitch')],axis=1)
data = pd.concat([data,pd.get_dummies(data["p_throws"], prefix='pitcher_throws')],axis=1)
data = pd.concat([data,pd.get_dummies(data["stand"], prefix='batter_stands')],axis=1)
data = pd.concat([data,pd.get_dummies(data["description"], prefix='prev_desc')],axis=1)
data = data.drop(['inning_topbot', 'p_throws', 'stand', "description", "prev_pitch"], axis = 1)

categoricalMask = data.dtypes==object
categoricalCols = data.columns[categoricalMask].tolist()
categoricalCols


le = LabelEncoder()
data[categoricalCols] = data[categoricalCols].apply(lambda col: le.fit_transform(col))
data = data.sample(frac=1).reset_index(drop=True)

features = data.columns.tolist()[1:]
X = data[features]
y = data["pitch_type"]

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, train_size=0.8)

print(y_train.value_counts())
smote = SMOTE("all")
X_train, y_train = smote.fit_sample(X_train, y_train)
print(y_train.value_counts())

num_classes=len(data["pitch_type"].unique())
y_train = to_categorical(y_train, num_classes)
y_test = to_categorical(y_test, num_classes)

def get_compiled_model():
  model = tf.keras.Sequential([
    tf.keras.layers.Dense(16, input_shape=(len(features),), activation='relu', name='fc1'), #layer 1
    tf.keras.layers.Dense(8, activation='relu', name='fc2'), #layer 2
    tf.keras.layers.Dense(num_classes, activation='softmax', name='output')
  ])

  model.compile(optimizer='adam',
                loss='binary_crossentropy',
                metrics=['accuracy'])
  return model

model = get_compiled_model()
model.fit(X_train, y_train, batch_size=100, epochs=200)

results = model.evaluate(X_test, y_test)

print('Final test set loss: {:4f}'.format(results[0]))
print('Final test set accuracy: {:4f}'.format(results[1]))

pitches = le.inverse_transform(data["pitch_type"].unique())

data["pitch_type"].unique()

some_data = data.sample(n=1)
ynew = model.predict_classes(some_data[features])
print(some_data)
print(le.inverse_transform(ynew))