onnx
(OneVsOneClassifier) Not able to convert sklearn model using pipeline to ONNX format for real time inferencing
It is a multi-class classification model with sklearn.
I am using OneVsOneClassifier model to train and predict 150 intents. Its a multi-class classification problem.
Data:
text intents
text1 int1
text2 int2
I convert these intents in labels using:
le = LabelEncoder()
y_train = le.fit_transform(y_train)
y_test = le.fit_transform(y_test)
Expectation:
Without changing the training pipeline or parameters, note the inference time. Currently, it's slow, ~1second for 1 inference. So to convert pipeline to ONNX format and then use for inferencing on 1 example.
Code:
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.compose import ColumnTransformer
from sklearn.multiclass import OneVsRestClassifier, OneVsOneClassifier
from sklearn.pipeline import Pipeline
from sklearn.svm import SVC,LinearSVC
def create_pipe(clf):
# Each pipeline uses the same column transformer.
column_trans = ColumnTransformer(
[('Text', TfidfVectorizer(), 'text')
],
remainder='drop')
pipeline = Pipeline([('prep',column_trans),
('clf', clf)])
return pipeline
def fit_and_print(pipeline):
pipeline.fit(X_train, y_train)
y_pred = pipeline.predict(X_test)
print(metrics.classification_report(y_test, y_pred,
target_names=le.classes_,
digits=3))
clf = OneVsOneClassifier(LinearSVC(random_state=42, class_weight='balanced'))
pipeline = create_pipe(clf)
%time fit_and_print(pipeline)
# convert input to df
def create_test_data(x):
d = {'text' : x}
df = pd.DataFrame(d, index=[0])
return df
revs=[]
for idx in [948, 5717, 458]:
cur = test.loc[idx, 'text']
revs.append(cur)
print(revs)
revs=sam['text'].values
%%time
for rev in revs:
c_res = pipeline.predict(create_test_data(rev))
print(rev, '=', labels[c_res[0]])
ONNX conversion code
from skl2onnx import convert_sklearn
from skl2onnx.common.data_types import FloatTensorType, StringTensorType
initial_type = [('UTTERANCE', StringTensorType([None, 2]))]
model_onnx = convert_sklearn(pipeline, initial_types=initial_type)
Error
MissingShapeCalculator: Unable to find a shape calculator for type '<class 'sklearn.multiclass.OneVsOneClassifier'>'.
It usually means the pipeline being converted contains a
transformer or a predictor with no corresponding converter
implemented in sklearn-onnx. If the converted is implemented
in another library, you need to register
the converted so that it can be used by sklearn-onnx (function
update_registered_converter). If the model is not yet covered
by sklearn-onnx, you may raise an issue to
https://github.com/onnx/sklearn-onnx/issues
to get the converter implemented or even contribute to the
project. If the model is a custom model, a new converter must
be implemented. Examples can be found in the gallery.
How to resolve this ? Also how to do prediction after converting to ONNX format?
The converter for OneVsOneClassifier has not been implemented yet. It should not be too complicated to do. In the meantime, OneVsRestClassifier has a converter.
The converter for OneVsOneClassifier has not been implemented yet. It should not be too complicated to do. In the meantime, OneVsRestClassifier has a converter.
But in the document OneVsOneClassifier was mentioned. Strange though. When can we expect to get it implemented?
@pratikchhapolika I'm working on this converter now. Can you please share me some of your data (X_train, y_train, X_test, y_test, etc.) so I can do a fully testing?
@pratikchhapolika I'm working on this converter now. Can you please share me some of your data (X_train, y_train, X_test, y_test, etc.) so I can do a fully testing?
sure. I will share it by end of day today.
@pratikchhapolika I'm working on this converter now. Can you please share me some of your data (X_train, y_train, X_test, y_test, etc.) so I can do a fully testing?
@xiaowuhu Please find the data-set
@pratikchhapolika I'm working on this converter now. Can you please share me some of your data (X_train, y_train, X_test, y_test, etc.) so I can do a fully testing?
@xiaowuhu
I am using OneVsOneClassifier with ColumnTransformer([('Text', TfidfVectorizer(), 'UTTERANCE')],remainder='drop')
def create_pipe(clf):
# Each pipeline uses the same column transformer.
column_trans = ColumnTransformer([('Text', TfidfVectorizer(), 'UTTERANCE')],remainder='drop')
pipeline = Pipeline([('prep',column_trans),
('clf', clf)])
return pipeline
clf = OneVsOneClassifier(LinearSVC(random_state=42, class_weight='balanced'))
pipeline = create_pipe(clf)
def fit_and_print(pipeline):
pipeline.fit(X_train, y_train)
y_pred = pipeline.predict(X_test)
print(metrics.classification_report(y_test, y_pred,
target_names=le.classes_,
digits=3))
%time fit_and_print(pipeline)
got it. will update you when I am ready.
sure @xiaowuhu Thank you!
@pratikchhapolika We finished the OvO converter, please pip install from GitHub to have a try. We didn't release it to PiPy yet.
@pratikchhapolika We finished the OvO converter, please pip install from GitHub to have a try. We didn't release it to PiPy yet.
@xiaowuhu Could you please show, how can I use it in my use-case or example above? Could you please post the code here?
@xiaowuhu Also please let us know how to install this particular ovo converter and use it.
@xiaowuhu Also please let us know how to install this particular
ovoconverter and use it.
pip install git+https://github.com/onnx/sklearn-onnx
if you can uninstall the previous package coming from PyPI, it will be more clear to check the above installation.
@pratikchhapolika We finished the OvO converter, please pip install from GitHub to have a try. We didn't release it to PiPy yet.
@xiaowuhu Could you please show, how can I use it in my use-case or example above? Could you please post the code here?
Please help with this also @xiaowuhu
@pratikchhapolika sorry I cannot make your sample (below code) working even in sk-learn, so cannot test the converter.
clf = OneVsOneClassifier(LinearSVC(random_state=42, class_weight='balanced'))
column_trans = ColumnTransformer([('Text', TfidfVectorizer(), 'UTTERANCE')], remainder='drop')
pipeline = Pipeline([('prep', column_trans), ('clf', clf)])
pipeline.fit(X_train, y_train)
y_pred = pipeline.predict(X_test)
print(y_pred)
return clf
we are using this test case: https://github.com/onnx/sklearn-onnx/blob/main/tests/test_sklearn_one_vs_one_classifier_converter.py
al_types=initial_type)
@xiaowuhu Thank you. Let me try on my notebook and check. Meanwhile is it fine to keep this open?
@pratikchhapolika
sorry to ask for the data format of X_train and others. Is it looks like pandas DataFrame:
or just a python list? my goal is to make below code work:clf = OneVsOneClassifier(LinearSVC(random_state=42, class_weight='balanced')) column_trans = ColumnTransformer([('Text', TfidfVectorizer(), 'UTTERANCE')], remainder='drop') pipeline = Pipeline([('prep', column_trans), ('clf', clf)]) pipeline.fit(X_train, y_train) y_pred = pipeline.predict(X_test) print(y_pred) return clf
Hi @xiaowuhu , could you please delete the X_train data from comment above. As it might violate some of my company's privacy.