pythonClient/pipeline/imdb.py at main · SGMPAndre/pythonClient · 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
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
from learning_orchestra_client.dataset.csv import DatasetCsv
from learning_orchestra_client.function.python import FunctionPython
from learning_orchestra_client.model.scikitlearn import ModelScikitLearn
from learning_orchestra_client.train.scikitlearn import TrainScikitLearn
from learning_orchestra_client.predict.scikitlearn import PredictScikitLearn

CLUSTER_IP = "http://34.123.167.241"

dataset_csv = DatasetCsv(CLUSTER_IP)

dataset_csv.insert_dataset_sync(
    dataset_name="sentiment_analysis",
    url="https://drive.google.com/u/0/uc?"
        "id=1PSLWHbKR_cuKvGKeOSl913kCfs-DJE2n&export=download",
)

function_python = FunctionPython(CLUSTER_IP)

explore_dataset = '''
pos=data[data["label"]=="1"]
neg=data[data["label"]=="0"]

total_rows = len(pos) + len(neg)

print("Positive = " + str(len(pos) / total_rows))
print("Negative = " + str(len(neg) / total_rows))

response = None
'''

function_python.run_function_sync(
    name="sentiment_analysis_exploring",
    parameters={"data": "$sentiment_analysis"},
    code=explore_dataset)

print(function_python.search_execution_content(
    name="sentiment_analysis_exploring",
    limit=1,
    skip=1,
    pretty_response=True))

dataset_preprocessing = '''
import re;


def preprocessor(text):
    global re
    text = re.sub("<[^>]*>", "", text)
    emojis = re.findall("(?::|;|=)(?:-)?(?:\)|\(|D|P)", text)
    text = re.sub("[\W]+", " ", text.lower()) + \
           " ".join(emojis).replace("-", "")
    return text


data["text"] = data["text"].apply(preprocessor)

from nltk.stem.porter import PorterStemmer

porter = PorterStemmer()


def tokenizer_porter(text):
    global porter
    return [porter.stem(word) for word in text.split()]


from sklearn.feature_extraction.text import TfidfVectorizer

tfidf = TfidfVectorizer(strip_accents=None,
                        lowercase=False,
                        preprocessor=None,
                        tokenizer=tokenizer_porter,
                        use_idf=True,
                        norm="l2",
                        smooth_idf=True)

y = data.label.values
x = tfidf.fit_transform(data.text)

from sklearn.model_selection import train_test_split

X_train, X_test, y_train, y_test = train_test_split(x, y,
                                                    random_state=1,
                                                    test_size=0.5,
                                                    shuffle=False)

response = {
    "X_train": X_train,
    "X_test": X_test,
    "y_train": y_train,
    "y_test": y_test
}
'''

function_python.run_function_sync(
    name="sentiment_analysis_preprocessed",
    parameters={
        "data": "$sentiment_analysis"
    },
    code=dataset_preprocessing
)

model_scikitlearn = ModelScikitLearn(CLUSTER_IP)

model_scikitlearn.create_model_sync(
    name="sentiment_analysis_logistic_regression_cv",
    module_path="sklearn.linear_model",
    class_name="LogisticRegressionCV",
    class_parameters={
        "cv": 5,
        "scoring": "accuracy",
        "random_state": 0,
        "n_jobs": -1,
        "verbose": 3,
        "max_iter": 100
    }

)

train_scikitlearn = TrainScikitLearn(CLUSTER_IP)
train_scikitlearn.create_training_sync(
    parent_name="sentiment_analysis_logistic_regression_cv",
    name="sentiment_analysis_logistic_regression_cv_trained",
    model_name="sentiment_analysis_logistic_regression_cv",
    method_name="fit",
    parameters={
        "X": "$sentiment_analysis_preprocessed.X_train",
        "y": "$sentiment_analysis_preprocessed.y_train",
    }
)

predict_scikitlearn = PredictScikitLearn(CLUSTER_IP)
predict_scikitlearn.create_prediction_sync(
    parent_name="sentiment_analysis_logistic_regression_cv_trained",
    name="sentiment_analysis_logistic_regression_cv_predicted",
    model_name="sentiment_analysis_logistic_regression_cv",
    method_name="predict",
    parameters={
        "X": "$sentiment_analysis_preprocessed.X_test",
    }

)

logistic_regression_cv_accuracy = '''
from sklearn import metrics

print("Accuracy: ",metrics.accuracy_score(y_test, y_pred))

response = None
'''
function_python.run_function_sync(
    name="sentiment_analysis_logistic_regression_cv_accuracy",
    parameters={
        "y_test": "$sentiment_analysis_preprocessed.y_test",
        "y_pred": "$sentiment_analysis_logistic_regression_cv_predicted"
    },
    code=logistic_regression_cv_accuracy
)

print(function_python.search_execution_content(
    name="sentiment_analysis_logistic_regression_cv_accuracy",
    pretty_response=True))