Explainable AI in Spanish

• Introducción
• Instalación
  • Carga de datos
  • Corregir datos desbalanceados
  • Dividir entre set de entrenamiento y prueba (test)
  • Funciones
  • Entrenar modelo Boosted Tree
  • Confusion matrix
  • Preparando el despliegue del modelo en AI Platform (con Explainability)
    • Exportamos el modelo en TensorFlow 1.x
  • Despliegue e AI Platform explanations
  • Predicción con explicabilidad
    • Paso 1: preparamos datos (data.txt)
    • Step 2: enviamos data.txt al modelo con gcloud beta ai-platform explain

Introducción

Aviso: la versión original en inglés de este notebook está en este Codelab

Instalación

import itertools
import numpy as np
import pandas as pd
import tensorflow as tf
import json
import matplotlib.pyplot as plt

from sklearn.utils import shuffle
from sklearn.metrics import confusion_matrix

Carga de datos

Es un dataset sintético de Kaggle, con trasacciones fraudulentas. Son 6.3 millones de registros, y 8000 son fraudulentas (sólo el 0.1!)

!gsutil cp gs://financial_fraud_detection/fraud_data_kaggle.csv .

Copying gs://financial_fraud_detection/fraud_data_kaggle.csv...
| [1 files][470.7 MiB/470.7 MiB]                                                
Operation completed over 1 objects/470.7 MiB.                                    

data = pd.read_csv('fraud_data_kaggle.csv')
data.head()
data.size

69988820

data['isFraud'].value_counts()

0    6354407
1       8213
Name: isFraud, dtype: int64

Corregir datos desbalanceados

Usamos DOWNSAMPLING: consiste en usar todos los fraudulentos (8000) y solo 0.005 (31000) de los no-fraudulentos (clase minoritaria)

fraud = data[data['isFraud'] == 1]
not_fraud = data[data['isFraud'] == 0]

# Take a random sample of non fraud rows
not_fraud_sample = not_fraud.sample(random_state=2, frac=.005)

# Put it back together and shuffle
df = pd.concat([not_fraud_sample,fraud])
df = shuffle(df, random_state=2)

# Remove a few columns (isFraud is the label column we'll use, not isFlaggedFraud)
df = df.drop(columns=['nameOrig', 'nameDest', 'isFlaggedFraud'])

# Preview the updated dataset
df.head()

	step	type	amount	oldbalanceOrg	newbalanceOrig	oldbalanceDest	newbalanceDest	isFraud
5777870	400	PAYMENT	65839.41	0.00	0.00	0.0	0.0	0
6362412	726	TRANSFER	561446.32	561446.32	0.00	0.0	0.0	1
5927827	404	PAYMENT	3828.08	10455.17	6627.09	0.0	0.0	0
5987904	410	TRANSFER	557950.06	557950.06	0.00	0.0	0.0	1
5706694	398	PAYMENT	1376.57	368349.14	366972.57	0.0	0.0	0

df['isFraud'].value_counts()

0    31772
1     8213
Name: isFraud, dtype: int64

Dividir entre set de entrenamiento y prueba (test)

train_test_split = int(len(df) * .8)

train_set = df[:train_test_split]
test_set = df[train_test_split:]

train_labels = train_set.pop('isFraud')
test_labels = test_set.pop('isFraud')

Definir features

fc = tf.feature_column
CATEGORICAL_COLUMNS = ['type']
NUMERIC_COLUMNS = ['step', 'amount', 'oldbalanceOrg', 'newbalanceOrig', 'oldbalanceDest', 'newbalanceDest']

def one_hot_cat_column(feature_name, vocab):
    return tf.feature_column.indicator_column(tf.feature_column.categorical_column_with_vocabulary_list(feature_name, vocab))

feature_columns = []

for feature_name in CATEGORICAL_COLUMNS:
    vocabulary = train_set[feature_name].unique()
    feature_columns.append(one_hot_cat_column(feature_name, vocabulary))

for feature_name in NUMERIC_COLUMNS:
  feature_columns.append(tf.feature_column.numeric_column(feature_name,
                                           dtype=tf.float32))

Funciones

NUM_EXAMPLES = len(train_labels)
def make_input_fn(X, y, n_epochs=None, shuffle=True):
  def input_fn():
    dataset = tf.data.Dataset.from_tensor_slices((dict(X), y))
    if shuffle:
      dataset = dataset.shuffle(NUM_EXAMPLES)
    dataset = dataset.repeat(n_epochs)
    dataset = dataset.batch(NUM_EXAMPLES)
    return dataset
  return input_fn

# Define training and evaluation input functions
train_input_fn = make_input_fn(train_set, train_labels)
eval_input_fn = make_input_fn(test_set, test_labels, shuffle=False, n_epochs=1)

Entrenar modelo Boosted Tree

n_batches = 1
model = tf.estimator.BoostedTreesClassifier(feature_columns,
                                          n_batches_per_layer=n_batches)

INFO:tensorflow:Using default config.
WARNING:tensorflow:Using temporary folder as model directory: /tmp/tmp8ko2kwtc
INFO:tensorflow:Using config: {'_model_dir': '/tmp/tmp8ko2kwtc', '_tf_random_seed': None, '_save_summary_steps': 100, '_save_checkpoints_steps': None, '_save_checkpoints_secs': 600, '_session_config': allow_soft_placement: true
graph_options {
  rewrite_options {
    meta_optimizer_iterations: ONE
  }
}
, '_keep_checkpoint_max': 5, '_keep_checkpoint_every_n_hours': 10000, '_log_step_count_steps': 100, '_train_distribute': None, '_device_fn': None, '_protocol': None, '_eval_distribute': None, '_experimental_distribute': None, '_experimental_max_worker_delay_secs': None, '_session_creation_timeout_secs': 7200, '_service': None, '_cluster_spec': <tensorflow.python.training.server_lib.ClusterSpec object at 0x7fcb4fb77810>, '_task_type': 'worker', '_task_id': 0, '_global_id_in_cluster': 0, '_master': '', '_evaluation_master': '', '_is_chief': True, '_num_ps_replicas': 0, '_num_worker_replicas': 1}
WARNING:tensorflow:From /opt/conda/lib/python3.7/site-packages/tensorflow_estimator/python/estimator/canned/boosted_trees.py:369: VocabularyListCategoricalColumn._num_buckets (from tensorflow.python.feature_column.feature_column_v2) is deprecated and will be removed in a future version.
Instructions for updating:
The old _FeatureColumn APIs are being deprecated. Please use the new FeatureColumn APIs instead.

model.train(train_input_fn, max_steps=100)

WARNING:tensorflow:From /opt/conda/lib/python3.7/site-packages/tensorflow_core/python/ops/resource_variable_ops.py:1630: calling BaseResourceVariable.__init__ (from tensorflow.python.ops.resource_variable_ops) with constraint is deprecated and will be removed in a future version.
Instructions for updating:
If using Keras pass *_constraint arguments to layers.
WARNING:tensorflow:From /opt/conda/lib/python3.7/site-packages/tensorflow_core/python/training/training_util.py:236: Variable.initialized_value (from tensorflow.python.ops.variables) is deprecated and will be removed in a future version.
Instructions for updating:
Use Variable.read_value. Variables in 2.X are initialized automatically both in eager and graph (inside tf.defun) contexts.
INFO:tensorflow:Calling model_fn.
WARNING:tensorflow:From /opt/conda/lib/python3.7/site-packages/tensorflow_core/python/feature_column/feature_column_v2.py:4271: IndicatorColumn._variable_shape (from tensorflow.python.feature_column.feature_column_v2) is deprecated and will be removed in a future version.
Instructions for updating:
The old _FeatureColumn APIs are being deprecated. Please use the new FeatureColumn APIs instead.
WARNING:tensorflow:From /opt/conda/lib/python3.7/site-packages/tensorflow_estimator/python/estimator/canned/boosted_trees.py:214: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.
Instructions for updating:
Use `tf.cast` instead.
WARNING:tensorflow:From /opt/conda/lib/python3.7/site-packages/tensorflow_estimator/python/estimator/canned/head.py:437: to_float (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.
Instructions for updating:
Use `tf.cast` instead.
WARNING:tensorflow:From /opt/conda/lib/python3.7/site-packages/tensorflow_core/python/ops/nn_impl.py:183: where (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.where in 2.0, which has the same broadcast rule as np.where
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Create CheckpointSaverHook.
WARNING:tensorflow:Issue encountered when serializing resources.
Type is unsupported, or the types of the items don't match field type in CollectionDef. Note this is a warning and probably safe to ignore.
'_Resource' object has no attribute 'name'
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
WARNING:tensorflow:Issue encountered when serializing resources.
Type is unsupported, or the types of the items don't match field type in CollectionDef. Note this is a warning and probably safe to ignore.
'_Resource' object has no attribute 'name'
INFO:tensorflow:Saving checkpoints for 0 into /tmp/tmp8ko2kwtc/model.ckpt.
WARNING:tensorflow:Issue encountered when serializing resources.
Type is unsupported, or the types of the items don't match field type in CollectionDef. Note this is a warning and probably safe to ignore.
'_Resource' object has no attribute 'name'
INFO:tensorflow:loss = 0.6931538, step = 0
WARNING:tensorflow:It seems that global step (tf.train.get_global_step) has not been increased. Current value (could be stable): 0 vs previous value: 0. You could increase the global step by passing tf.train.get_global_step() to Optimizer.apply_gradients or Optimizer.minimize.
INFO:tensorflow:global_step/sec: 1.56855
INFO:tensorflow:loss = 0.022666413, step = 99 (63.755 sec)
INFO:tensorflow:Saving checkpoints for 100 into /tmp/tmp8ko2kwtc/model.ckpt.
WARNING:tensorflow:Issue encountered when serializing resources.
Type is unsupported, or the types of the items don't match field type in CollectionDef. Note this is a warning and probably safe to ignore.
'_Resource' object has no attribute 'name'
INFO:tensorflow:Loss for final step: 0.022666413.

<tensorflow_estimator.python.estimator.canned.boosted_trees.BoostedTreesClassifier at 0x7fcb4fb71dd0>

result = model.evaluate(eval_input_fn)
print(pd.Series(result))

INFO:tensorflow:Calling model_fn.
WARNING:tensorflow:From /opt/conda/lib/python3.7/site-packages/tensorflow_core/python/ops/metrics_impl.py:2026: div (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.
Instructions for updating:
Deprecated in favor of operator or tf.math.divide.
WARNING:tensorflow:Trapezoidal rule is known to produce incorrect PR-AUCs; please switch to "careful_interpolation" instead.
WARNING:tensorflow:Trapezoidal rule is known to produce incorrect PR-AUCs; please switch to "careful_interpolation" instead.
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Starting evaluation at 2020-06-22T08:11:13Z
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from /tmp/tmp8ko2kwtc/model.ckpt-100
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Finished evaluation at 2020-06-22-08:11:15
INFO:tensorflow:Saving dict for global step 100: accuracy = 0.9949981, accuracy_baseline = 0.8006753, auc = 0.99874365, auc_precision_recall = 0.99814403, average_loss = 0.024392342, global_step = 100, label/mean = 0.19932474, loss = 0.024392342, precision = 0.97903824, prediction/mean = 0.2005636, recall = 0.9962359
WARNING:tensorflow:Issue encountered when serializing resources.
Type is unsupported, or the types of the items don't match field type in CollectionDef. Note this is a warning and probably safe to ignore.
'_Resource' object has no attribute 'name'
INFO:tensorflow:Saving 'checkpoint_path' summary for global step 100: /tmp/tmp8ko2kwtc/model.ckpt-100
accuracy                  0.994998
accuracy_baseline         0.800675
auc                       0.998744
auc_precision_recall      0.998144
average_loss              0.024392
label/mean                0.199325
loss                      0.024392
precision                 0.979038
prediction/mean           0.200564
recall                    0.996236
global_step             100.000000
dtype: float64

pred_dicts = list(model.predict(eval_input_fn))
probabilities = pd.Series([pred['logistic'][0] for pred in pred_dicts])

for i,val in enumerate(probabilities[:30]):
  print('Predicted: ', round(val), 'Actual: ', test_labels.iloc[i])
  print()

INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from /tmp/tmp8ko2kwtc/model.ckpt-100
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
Predicted:  0 Actual:  0

Predicted:  1 Actual:  1

Predicted:  0 Actual:  0

Predicted:  0 Actual:  0

Predicted:  0 Actual:  0

Predicted:  0 Actual:  0

Predicted:  0 Actual:  0

Predicted:  0 Actual:  0

Predicted:  0 Actual:  0

Predicted:  0 Actual:  0

Predicted:  0 Actual:  0

Predicted:  0 Actual:  0

Predicted:  0 Actual:  0

Predicted:  0 Actual:  0

Predicted:  0 Actual:  0

Predicted:  0 Actual:  0

Predicted:  0 Actual:  0

Predicted:  0 Actual:  0

Predicted:  0 Actual:  0

Predicted:  1 Actual:  1

Predicted:  0 Actual:  0

Predicted:  0 Actual:  0

Predicted:  0 Actual:  0

Predicted:  0 Actual:  0

Predicted:  1 Actual:  1

Predicted:  0 Actual:  0

Predicted:  0 Actual:  0

Predicted:  0 Actual:  0

Predicted:  0 Actual:  0

Predicted:  0 Actual:  0

Confusion matrix

y_pred = []

for i in probabilities.values:
  y_pred.append(int(round(i)))

cm = confusion_matrix(test_labels.values, y_pred)
print(cm)

[[6369   34]
 [   6 1588]]

def plot_confusion_matrix(cm, classes,
                          normalize=False,
                          title='Confusion matrix',
                          cmap=plt.cm.Blues):
    """
    This function prints and plots the confusion matrix.
    Normalization can be applied by setting `normalize=True`.
    """
    plt.imshow(cm, interpolation='nearest', cmap=cmap)
    plt.title(title)
    plt.colorbar()
    tick_marks = np.arange(len(classes))
    plt.xticks(tick_marks, classes, rotation=45)
    plt.yticks(tick_marks, classes)

    if normalize:
        cm = np.round(cm.astype('float') / cm.sum(axis=1)[:, np.newaxis], 3)

    thresh = cm.max() / 2.
    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):
        plt.text(j, i, cm[i, j],
                 horizontalalignment="center",
                 color="white" if cm[i, j] > thresh else "black")

    plt.tight_layout()
    plt.ylabel('True label')
    plt.xlabel('Predicted label')

classes = ['not fraud', 'fraud']
plot_confusion_matrix(cm, classes, normalize=True)

Preparando el despliegue del modelo en AI Platform (con Explainability)

GCP_PROJECT = 'windy-site-254307'
MODEL_BUCKET = 'gs://fraud-detection-explainable-ai'
!gsutil mb $MODEL_BUCKET

Creating gs://fraud-detection-explainable-ai/...
ServiceException: 409 Bucket fraud-detection-explainable-ai already exists.

Exportamos el modelo en TensorFlow 1.x

import tensorflow.compat.v1 as tf
tf.disable_v2_behavior()
def json_serving_input_fn():
  inputs = {}
  for feat in feature_columns:
      if feat.name == "type_indicator":
            inputs['type'] = tf.placeholder(shape=[None], name=feat.name, dtype=tf.string)
      else:
          inputs[feat.name] = tf.placeholder(shape=[None], name=feat.name, dtype=feat.dtype)
  return tf.estimator.export.ServingInputReceiver(inputs, inputs)

export_path = model.export_saved_model(
    MODEL_BUCKET + '/explanations',
    serving_input_receiver_fn=json_serving_input_fn
).decode('utf-8')

tf.enable_v2_behavior()

!saved_model_cli show --dir $export_path --all

MetaGraphDef with tag-set: 'serve' contains the following SignatureDefs:

signature_def['predict']:
  The given SavedModel SignatureDef contains the following input(s):
    inputs['amount'] tensor_info:
        dtype: DT_FLOAT
        shape: (-1)
        name: amount:0
    inputs['newbalanceDest'] tensor_info:
        dtype: DT_FLOAT
        shape: (-1)
        name: newbalanceDest:0
    inputs['newbalanceOrig'] tensor_info:
        dtype: DT_FLOAT
        shape: (-1)
        name: newbalanceOrig:0
    inputs['oldbalanceDest'] tensor_info:
        dtype: DT_FLOAT
        shape: (-1)
        name: oldbalanceDest:0
    inputs['oldbalanceOrg'] tensor_info:
        dtype: DT_FLOAT
        shape: (-1)
        name: oldbalanceOrg:0
    inputs['step'] tensor_info:
        dtype: DT_FLOAT
        shape: (-1)
        name: step:0
    inputs['type'] tensor_info:
        dtype: DT_STRING
        shape: (-1)
        name: type_indicator:0
  The given SavedModel SignatureDef contains the following output(s):
    outputs['all_class_ids'] tensor_info:
        dtype: DT_INT32
        shape: (-1, 2)
        name: boosted_trees/head/predictions/Tile:0
    outputs['all_classes'] tensor_info:
        dtype: DT_STRING
        shape: (-1, 2)
        name: boosted_trees/head/predictions/Tile_1:0
    outputs['class_ids'] tensor_info:
        dtype: DT_INT64
        shape: (-1, 1)
        name: boosted_trees/head/predictions/ExpandDims:0
    outputs['classes'] tensor_info:
        dtype: DT_STRING
        shape: (-1, 1)
        name: boosted_trees/head/predictions/str_classes:0
    outputs['logistic'] tensor_info:
        dtype: DT_FLOAT
        shape: (-1, 1)
        name: boosted_trees/head/predictions/logistic:0
    outputs['logits'] tensor_info:
        dtype: DT_FLOAT
        shape: (-1, 1)
        name: boosted_trees/BoostedTreesPredict:0
    outputs['probabilities'] tensor_info:
        dtype: DT_FLOAT
        shape: (-1, 2)
        name: boosted_trees/head/predictions/probabilities:0
  Method name is: tensorflow/serving/predict

not_fraud.median()

step                 239.00
amount             74684.72
oldbalanceOrg      14069.00
newbalanceOrig         0.00
oldbalanceDest    133311.80
newbalanceDest    214881.70
isFraud                0.00
isFlaggedFraud         0.00
dtype: float64

not_fraud['type'].value_counts()

CASH_OUT    2233384
PAYMENT     2151495
CASH_IN     1399284
TRANSFER     528812
DEBIT         41432
Name: type, dtype: int64

!gsutil cp explanation_metadata.json $export_path

Copying file://explanation_metadata.json [Content-Type=application/json]...
/ [1 files][  718.0 B/  718.0 B]                                                
Operation completed over 1 objects/718.0 B.                                      

!gsutil cp gs://fraud-detection-explainable-ai/explanations/1592218671/explanation_metadata.json .
!gsutil cp explanation_metadata.json $export_path

Despliegue e AI Platform explanations

MODEL = 'fraud_detection_4'
VERSION = 'v3'

!gcloud ai-platform models create $MODEL

WARNING: Using endpoint [https://ml.googleapis.com/]
WARNING: Please explicitly specify a region. Using [us-central1] by default on https://ml.googleapis.com. Please note that your model will be inaccessible from https://us-central1-ml.googelapis.com

Learn more about regional endpoints and see a list of available regions: https://cloud.google.com/ai-platform/prediction/docs/regional-endpoints
ERROR: (gcloud.ai-platform.models.create) Resource in project [windy-site-254307] is the subject of a conflict: Field: model.name Error: A model with the same name already exists.
- '@type': type.googleapis.com/google.rpc.BadRequest
  fieldViolations:
  - description: A model with the same name already exists.
    field: model.name

!gcloud beta ai-platform versions create v3 \
--model $MODEL \
--origin $export_path \
--runtime-version 1.15 \
--framework TENSORFLOW \
--python-version 3.7 \
--machine-type n1-standard-4 \
--explanation-method 'sampled-shapley' \
--num-paths 10

WARNING: Using endpoint [https://ml.googleapis.com/]
Explanations reflect patterns in your model, but don't necessarily reveal fundamental relationships about your data population. See https://cloud.google.com/ml-engine/docs/ai-explanations/limitations for more information.
ERROR: (gcloud.beta.ai-platform.versions.create) ALREADY_EXISTS: Field: version.name Error: A version with the same name already exists.
- '@type': type.googleapis.com/google.rpc.BadRequest
  fieldViolations:
  - description: A version with the same name already exists.
    field: version.name

!gcloud ai-platform versions describe $VERSION --model $MODEL

WARNING: Using endpoint [https://ml.googleapis.com/]
createTime: '2020-06-15T14:20:08Z'
deploymentUri: gs://fraud-detection-explainable-ai/explanations/1592230714
etag: yuDcVfyBSAQ=
explanationConfig:
  sampledShapleyAttribution:
    numPaths: 10
framework: TENSORFLOW
isDefault: true
lastUseTime: '2020-06-18T07:20:48Z'
machineType: n1-standard-4
name: projects/windy-site-254307/models/fraud_detection_4/versions/v3
pythonVersion: '3.7'
runtimeVersion: '1.15'
state: READY

Predicción con explicabilidad

Paso 1: preparamos datos (data.txt)

fraud_indices = []

for i,val in enumerate(test_labels):
    if val == 1:
        fraud_indices.append(i)

num_test_examples = 5
import numpy as np 

def convert(o):
    if isinstance(o, np.generic): return o.item()  
    raise TypeError

for i in range(num_test_examples):
    test_json = {}
    ex = test_set.iloc[fraud_indices[i]]
    keys = ex.keys().tolist()
    vals = ex.values.tolist()
    for idx in range(len(keys)):
        test_json[keys[idx]] = vals[idx]

    print(test_json)
    with open('data.txt', 'a') as outfile:
        json.dump(test_json, outfile, default=convert)
        outfile.write('\n')

{'step': 476, 'type': 'TRANSFER', 'amount': 1048.63, 'oldbalanceOrg': 1048.63, 'newbalanceOrig': 0.0, 'oldbalanceDest': 0.0, 'newbalanceDest': 0.0}
{'step': 390, 'type': 'TRANSFER', 'amount': 638693.49, 'oldbalanceOrg': 638693.49, 'newbalanceOrig': 0.0, 'oldbalanceDest': 0.0, 'newbalanceDest': 0.0}
{'step': 355, 'type': 'CASH_OUT', 'amount': 5338162.8, 'oldbalanceOrg': 5338162.8, 'newbalanceOrig': 0.0, 'oldbalanceDest': 181895.58, 'newbalanceDest': 5520058.37}
{'step': 356, 'type': 'TRANSFER', 'amount': 357226.8, 'oldbalanceOrg': 357226.8, 'newbalanceOrig': 0.0, 'oldbalanceDest': 0.0, 'newbalanceDest': 0.0}
{'step': 345, 'type': 'TRANSFER', 'amount': 128936.95, 'oldbalanceOrg': 128936.95, 'newbalanceOrig': 0.0, 'oldbalanceDest': 0.0, 'newbalanceDest': 0.0}

!cat data.txt

{"step": 476, "type": "TRANSFER", "amount": 1048.63, "oldbalanceOrg": 1048.63, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 390, "type": "TRANSFER", "amount": 638693.49, "oldbalanceOrg": 638693.49, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 355, "type": "CASH_OUT", "amount": 5338162.8, "oldbalanceOrg": 5338162.8, "newbalanceOrig": 0.0, "oldbalanceDest": 181895.58, "newbalanceDest": 5520058.37}
{"step": 356, "type": "TRANSFER", "amount": 357226.8, "oldbalanceOrg": 357226.8, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 345, "type": "TRANSFER", "amount": 128936.95, "oldbalanceOrg": 128936.95, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 476, "type": "TRANSFER", "amount": 1048.63, "oldbalanceOrg": 1048.63, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 390, "type": "TRANSFER", "amount": 638693.49, "oldbalanceOrg": 638693.49, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 355, "type": "CASH_OUT", "amount": 5338162.8, "oldbalanceOrg": 5338162.8, "newbalanceOrig": 0.0, "oldbalanceDest": 181895.58, "newbalanceDest": 5520058.37}
{"step": 356, "type": "TRANSFER", "amount": 357226.8, "oldbalanceOrg": 357226.8, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 345, "type": "TRANSFER", "amount": 128936.95, "oldbalanceOrg": 128936.95, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 476, "type": "TRANSFER", "amount": 1048.63, "oldbalanceOrg": 1048.63, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 390, "type": "TRANSFER", "amount": 638693.49, "oldbalanceOrg": 638693.49, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 355, "type": "CASH_OUT", "amount": 5338162.8, "oldbalanceOrg": 5338162.8, "newbalanceOrig": 0.0, "oldbalanceDest": 181895.58, "newbalanceDest": 5520058.37}
{"step": 356, "type": "TRANSFER", "amount": 357226.8, "oldbalanceOrg": 357226.8, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 345, "type": "TRANSFER", "amount": 128936.95, "oldbalanceOrg": 128936.95, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 476, "type": "TRANSFER", "amount": 1048.63, "oldbalanceOrg": 1048.63, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 390, "type": "TRANSFER", "amount": 638693.49, "oldbalanceOrg": 638693.49, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 355, "type": "CASH_OUT", "amount": 5338162.8, "oldbalanceOrg": 5338162.8, "newbalanceOrig": 0.0, "oldbalanceDest": 181895.58, "newbalanceDest": 5520058.37}
{"step": 356, "type": "TRANSFER", "amount": 357226.8, "oldbalanceOrg": 357226.8, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 345, "type": "TRANSFER", "amount": 128936.95, "oldbalanceOrg": 128936.95, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 476, "type": "TRANSFER", "amount": 1048.63, "oldbalanceOrg": 1048.63, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 390, "type": "TRANSFER", "amount": 638693.49, "oldbalanceOrg": 638693.49, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 355, "type": "CASH_OUT", "amount": 5338162.8, "oldbalanceOrg": 5338162.8, "newbalanceOrig": 0.0, "oldbalanceDest": 181895.58, "newbalanceDest": 5520058.37}
{"step": 356, "type": "TRANSFER", "amount": 357226.8, "oldbalanceOrg": 357226.8, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 345, "type": "TRANSFER", "amount": 128936.95, "oldbalanceOrg": 128936.95, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 476, "type": "TRANSFER", "amount": 1048.63, "oldbalanceOrg": 1048.63, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 390, "type": "TRANSFER", "amount": 638693.49, "oldbalanceOrg": 638693.49, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 355, "type": "CASH_OUT", "amount": 5338162.8, "oldbalanceOrg": 5338162.8, "newbalanceOrig": 0.0, "oldbalanceDest": 181895.58, "newbalanceDest": 5520058.37}
{"step": 356, "type": "TRANSFER", "amount": 357226.8, "oldbalanceOrg": 357226.8, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 345, "type": "TRANSFER", "amount": 128936.95, "oldbalanceOrg": 128936.95, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 476, "type": "TRANSFER", "amount": 1048.63, "oldbalanceOrg": 1048.63, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 390, "type": "TRANSFER", "amount": 638693.49, "oldbalanceOrg": 638693.49, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 355, "type": "CASH_OUT", "amount": 5338162.8, "oldbalanceOrg": 5338162.8, "newbalanceOrig": 0.0, "oldbalanceDest": 181895.58, "newbalanceDest": 5520058.37}
{"step": 356, "type": "TRANSFER", "amount": 357226.8, "oldbalanceOrg": 357226.8, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 345, "type": "TRANSFER", "amount": 128936.95, "oldbalanceOrg": 128936.95, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 476, "type": "TRANSFER", "amount": 1048.63, "oldbalanceOrg": 1048.63, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 390, "type": "TRANSFER", "amount": 638693.49, "oldbalanceOrg": 638693.49, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 355, "type": "CASH_OUT", "amount": 5338162.8, "oldbalanceOrg": 5338162.8, "newbalanceOrig": 0.0, "oldbalanceDest": 181895.58, "newbalanceDest": 5520058.37}
{"step": 356, "type": "TRANSFER", "amount": 357226.8, "oldbalanceOrg": 357226.8, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 345, "type": "TRANSFER", "amount": 128936.95, "oldbalanceOrg": 128936.95, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 476, "type": "TRANSFER", "amount": 1048.63, "oldbalanceOrg": 1048.63, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 390, "type": "TRANSFER", "amount": 638693.49, "oldbalanceOrg": 638693.49, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 355, "type": "CASH_OUT", "amount": 5338162.8, "oldbalanceOrg": 5338162.8, "newbalanceOrig": 0.0, "oldbalanceDest": 181895.58, "newbalanceDest": 5520058.37}
{"step": 356, "type": "TRANSFER", "amount": 357226.8, "oldbalanceOrg": 357226.8, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}
{"step": 345, "type": "TRANSFER", "amount": 128936.95, "oldbalanceOrg": 128936.95, "newbalanceOrig": 0.0, "oldbalanceDest": 0.0, "newbalanceDest": 0.0}

Step 2: enviamos data.txt al modelo con gcloud beta ai-platform explain

explanations = !gcloud beta ai-platform explain --model $MODEL --version $VERSION --json-instances='data.txt' --verbosity error

'{   "explanations": [     {       "attributions_by_label": [         {           "approx_error": 0.07222782395555333,           "attributions": {             "amount": 0.6966777056455612,             "newbalanceDest": 0.1639272928237915,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.16475289762020112,             "oldbalanceOrg": -0.15600235760211945,             "step": 0.1192602276802063,             "type": -0.027230754494667053           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9744548797607422,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.004716743469869621,           "attributions": {             "amount": -0.0019459187984466552,             "newbalanceDest": 0.007967007160186768,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.008247834444046021,             "oldbalanceOrg": 0.9784113168716431,             "step": 0.0,             "type": -0.007150697708129883           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9985994100570679,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.0024439054423102693,           "attributions": {             "amount": -0.002712637186050415,             "newbalanceDest": -0.0005802184343338013,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.0,             "oldbalanceOrg": 0.9835709899663925,             "step": 0.0,             "type": 0.0           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9933480024337769,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.014761328724187625,           "attributions": {             "amount": -0.02181842029094696,             "newbalanceDest": 0.029705092310905457,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.038540315628051755,             "oldbalanceOrg": 0.9493075281381607,             "step": 0.0,             "type": -0.011098328232765197           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9977060556411743,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.036956189992043476,           "attributions": {             "amount": -0.05543297529220581,             "newbalanceDest": 0.09204374849796296,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.09319761395454407,             "oldbalanceOrg": 0.8222974985837936,             "step": 0.0,             "type": 0.028654450178146364           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9938302040100098,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.07222782395555333,           "attributions": {             "amount": 0.6966777056455612,             "newbalanceDest": 0.1639272928237915,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.16475289762020112,             "oldbalanceOrg": -0.15600235760211945,             "step": 0.1192602276802063,             "type": -0.027230754494667053           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9744548797607422,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.004716743469869621,           "attributions": {             "amount": -0.0019459187984466552,             "newbalanceDest": 0.007967007160186768,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.008247834444046021,             "oldbalanceOrg": 0.9784113168716431,             "step": 0.0,             "type": -0.007150697708129883           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9985994100570679,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.0024439054423102693,           "attributions": {             "amount": -0.002712637186050415,             "newbalanceDest": -0.0005802184343338013,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.0,             "oldbalanceOrg": 0.9835709899663925,             "step": 0.0,             "type": 0.0           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9933480024337769,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.014761328724187625,           "attributions": {             "amount": -0.02181842029094696,             "newbalanceDest": 0.029705092310905457,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.038540315628051755,             "oldbalanceOrg": 0.9493075281381607,             "step": 0.0,             "type": -0.011098328232765197           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9977060556411743,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.036956189992043476,           "attributions": {             "amount": -0.05543297529220581,             "newbalanceDest": 0.09204374849796296,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.09319761395454407,             "oldbalanceOrg": 0.8222974985837936,             "step": 0.0,             "type": 0.028654450178146364           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9938302040100098,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.07222782395555333,           "attributions": {             "amount": 0.6966777056455612,             "newbalanceDest": 0.1639272928237915,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.16475289762020112,             "oldbalanceOrg": -0.15600235760211945,             "step": 0.1192602276802063,             "type": -0.027230754494667053           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9744548797607422,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.004716743469869621,           "attributions": {             "amount": -0.0019459187984466552,             "newbalanceDest": 0.007967007160186768,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.008247834444046021,             "oldbalanceOrg": 0.9784113168716431,             "step": 0.0,             "type": -0.007150697708129883           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9985994100570679,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.0024439054423102693,           "attributions": {             "amount": -0.002712637186050415,             "newbalanceDest": -0.0005802184343338013,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.0,             "oldbalanceOrg": 0.9835709899663925,             "step": 0.0,             "type": 0.0           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9933480024337769,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.014761328724187625,           "attributions": {             "amount": -0.02181842029094696,             "newbalanceDest": 0.029705092310905457,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.038540315628051755,             "oldbalanceOrg": 0.9493075281381607,             "step": 0.0,             "type": -0.011098328232765197           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9977060556411743,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.036956189992043476,           "attributions": {             "amount": -0.05543297529220581,             "newbalanceDest": 0.09204374849796296,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.09319761395454407,             "oldbalanceOrg": 0.8222974985837936,             "step": 0.0,             "type": 0.028654450178146364           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9938302040100098,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.07222782395555333,           "attributions": {             "amount": 0.6966777056455612,             "newbalanceDest": 0.1639272928237915,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.16475289762020112,             "oldbalanceOrg": -0.15600235760211945,             "step": 0.1192602276802063,             "type": -0.027230754494667053           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9744548797607422,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.004716743469869621,           "attributions": {             "amount": -0.0019459187984466552,             "newbalanceDest": 0.007967007160186768,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.008247834444046021,             "oldbalanceOrg": 0.9784113168716431,             "step": 0.0,             "type": -0.007150697708129883           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9985994100570679,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.0024439054423102693,           "attributions": {             "amount": -0.002712637186050415,             "newbalanceDest": -0.0005802184343338013,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.0,             "oldbalanceOrg": 0.9835709899663925,             "step": 0.0,             "type": 0.0           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9933480024337769,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.014761328724187625,           "attributions": {             "amount": -0.02181842029094696,             "newbalanceDest": 0.029705092310905457,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.038540315628051755,             "oldbalanceOrg": 0.9493075281381607,             "step": 0.0,             "type": -0.011098328232765197           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9977060556411743,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.036956189992043476,           "attributions": {             "amount": -0.05543297529220581,             "newbalanceDest": 0.09204374849796296,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.09319761395454407,             "oldbalanceOrg": 0.8222974985837936,             "step": 0.0,             "type": 0.028654450178146364           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9938302040100098,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.07222782395555333,           "attributions": {             "amount": 0.6966777056455612,             "newbalanceDest": 0.1639272928237915,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.16475289762020112,             "oldbalanceOrg": -0.15600235760211945,             "step": 0.1192602276802063,             "type": -0.027230754494667053           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9744548797607422,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.004716743469869621,           "attributions": {             "amount": -0.0019459187984466552,             "newbalanceDest": 0.007967007160186768,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.008247834444046021,             "oldbalanceOrg": 0.9784113168716431,             "step": 0.0,             "type": -0.007150697708129883           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9985994100570679,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.0024439054423102693,           "attributions": {             "amount": -0.002712637186050415,             "newbalanceDest": -0.0005802184343338013,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.0,             "oldbalanceOrg": 0.9835709899663925,             "step": 0.0,             "type": 0.0           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9933480024337769,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.014761328724187625,           "attributions": {             "amount": -0.02181842029094696,             "newbalanceDest": 0.029705092310905457,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.038540315628051755,             "oldbalanceOrg": 0.9493075281381607,             "step": 0.0,             "type": -0.011098328232765197           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9977060556411743,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.036956189992043476,           "attributions": {             "amount": -0.05543297529220581,             "newbalanceDest": 0.09204374849796296,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.09319761395454407,             "oldbalanceOrg": 0.8222974985837936,             "step": 0.0,             "type": 0.028654450178146364           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9938302040100098,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.07222782395555333,           "attributions": {             "amount": 0.6966777056455612,             "newbalanceDest": 0.1639272928237915,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.16475289762020112,             "oldbalanceOrg": -0.15600235760211945,             "step": 0.1192602276802063,             "type": -0.027230754494667053           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9744548797607422,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.004716743469869621,           "attributions": {             "amount": -0.0019459187984466552,             "newbalanceDest": 0.007967007160186768,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.008247834444046021,             "oldbalanceOrg": 0.9784113168716431,             "step": 0.0,             "type": -0.007150697708129883           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9985994100570679,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.0024439054423102693,           "attributions": {             "amount": -0.002712637186050415,             "newbalanceDest": -0.0005802184343338013,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.0,             "oldbalanceOrg": 0.9835709899663925,             "step": 0.0,             "type": 0.0           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9933480024337769,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.014761328724187625,           "attributions": {             "amount": -0.02181842029094696,             "newbalanceDest": 0.029705092310905457,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.038540315628051755,             "oldbalanceOrg": 0.9493075281381607,             "step": 0.0,             "type": -0.011098328232765197           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9977060556411743,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.036956189992043476,           "attributions": {             "amount": -0.05543297529220581,             "newbalanceDest": 0.09204374849796296,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.09319761395454407,             "oldbalanceOrg": 0.8222974985837936,             "step": 0.0,             "type": 0.028654450178146364           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9938302040100098,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.07222782395555333,           "attributions": {             "amount": 0.6966777056455612,             "newbalanceDest": 0.1639272928237915,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.16475289762020112,             "oldbalanceOrg": -0.15600235760211945,             "step": 0.1192602276802063,             "type": -0.027230754494667053           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9744548797607422,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.004716743469869621,           "attributions": {             "amount": -0.0019459187984466552,             "newbalanceDest": 0.007967007160186768,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.008247834444046021,             "oldbalanceOrg": 0.9784113168716431,             "step": 0.0,             "type": -0.007150697708129883           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9985994100570679,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.0024439054423102693,           "attributions": {             "amount": -0.002712637186050415,             "newbalanceDest": -0.0005802184343338013,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.0,             "oldbalanceOrg": 0.9835709899663925,             "step": 0.0,             "type": 0.0           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9933480024337769,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.014761328724187625,           "attributions": {             "amount": -0.02181842029094696,             "newbalanceDest": 0.029705092310905457,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.038540315628051755,             "oldbalanceOrg": 0.9493075281381607,             "step": 0.0,             "type": -0.011098328232765197           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9977060556411743,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.036956189992043476,           "attributions": {             "amount": -0.05543297529220581,             "newbalanceDest": 0.09204374849796296,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.09319761395454407,             "oldbalanceOrg": 0.8222974985837936,             "step": 0.0,             "type": 0.028654450178146364           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9938302040100098,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.07222782395555333,           "attributions": {             "amount": 0.6966777056455612,             "newbalanceDest": 0.1639272928237915,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.16475289762020112,             "oldbalanceOrg": -0.15600235760211945,             "step": 0.1192602276802063,             "type": -0.027230754494667053           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9744548797607422,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.004716743469869621,           "attributions": {             "amount": -0.0019459187984466552,             "newbalanceDest": 0.007967007160186768,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.008247834444046021,             "oldbalanceOrg": 0.9784113168716431,             "step": 0.0,             "type": -0.007150697708129883           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9985994100570679,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.0024439054423102693,           "attributions": {             "amount": -0.002712637186050415,             "newbalanceDest": -0.0005802184343338013,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.0,             "oldbalanceOrg": 0.9835709899663925,             "step": 0.0,             "type": 0.0           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9933480024337769,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.014761328724187625,           "attributions": {             "amount": -0.02181842029094696,             "newbalanceDest": 0.029705092310905457,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.038540315628051755,             "oldbalanceOrg": 0.9493075281381607,             "step": 0.0,             "type": -0.011098328232765197           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9977060556411743,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.036956189992043476,           "attributions": {             "amount": -0.05543297529220581,             "newbalanceDest": 0.09204374849796296,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.09319761395454407,             "oldbalanceOrg": 0.8222974985837936,             "step": 0.0,             "type": 0.028654450178146364           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9938302040100098,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.07222782395555333,           "attributions": {             "amount": 0.6966777056455612,             "newbalanceDest": 0.1639272928237915,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.16475289762020112,             "oldbalanceOrg": -0.15600235760211945,             "step": 0.1192602276802063,             "type": -0.027230754494667053           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9744548797607422,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.004716743469869621,           "attributions": {             "amount": -0.0019459187984466552,             "newbalanceDest": 0.007967007160186768,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.008247834444046021,             "oldbalanceOrg": 0.9784113168716431,             "step": 0.0,             "type": -0.007150697708129883           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9985994100570679,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.0024439054423102693,           "attributions": {             "amount": -0.002712637186050415,             "newbalanceDest": -0.0005802184343338013,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.0,             "oldbalanceOrg": 0.9835709899663925,             "step": 0.0,             "type": 0.0           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9933480024337769,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.014761328724187625,           "attributions": {             "amount": -0.02181842029094696,             "newbalanceDest": 0.029705092310905457,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.038540315628051755,             "oldbalanceOrg": 0.9493075281381607,             "step": 0.0,             "type": -0.011098328232765197           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9977060556411743,           "label_index": 0,           "output_name": "prob"         }       ]     },     {       "attributions_by_label": [         {           "approx_error": 0.036956189992043476,           "attributions": {             "amount": -0.05543297529220581,             "newbalanceDest": 0.09204374849796296,             "newbalanceOrig": 0.0,             "oldbalanceDest": 0.09319761395454407,             "oldbalanceOrg": 0.8222974985837936,             "step": 0.0,             "type": 0.028654450178146364           },           "baseline_score": 0.013069868087768555,           "example_score": 0.9938302040100098,           "label_index": 0,           "output_name": "prob"         }       ]     }   ] }'

explain_dict = json.loads(explanations.s)

### Step 3: análisis de datos (notad el baseline)

print('Model baseline for fraud cases: ', explain_dict['explanations'][0]['attributions_by_label'][0]['baseline_score'], '\n')

Model baseline for fraud cases:  0.013069868087768555 

for i in explain_dict['explanations']:
    prediction_score = i['attributions_by_label'][0]['example_score']
    attributions = i['attributions_by_label'][0]['attributions']
    print('Model prediction:', prediction_score)
    fig, ax = plt.subplots()
    ax.barh(list(attributions.keys()), list(attributions.values()), align='center')
    plt.show()

Model prediction: 0.9744548797607422

Model prediction: 0.9985994100570679

Model prediction: 0.9933480024337769

Model prediction: 0.9977060556411743

Model prediction: 0.9938302040100098

Model prediction: 0.9744548797607422

Model prediction: 0.9985994100570679

Model prediction: 0.9933480024337769

Model prediction: 0.9977060556411743

Model prediction: 0.9938302040100098

Model prediction: 0.9744548797607422

Model prediction: 0.9985994100570679

Model prediction: 0.9933480024337769

Model prediction: 0.9977060556411743

Model prediction: 0.9938302040100098

Model prediction: 0.9744548797607422

Model prediction: 0.9985994100570679

Model prediction: 0.9933480024337769

Model prediction: 0.9977060556411743

Model prediction: 0.9938302040100098

Model prediction: 0.9744548797607422

Model prediction: 0.9985994100570679

Model prediction: 0.9933480024337769

Model prediction: 0.9977060556411743

Model prediction: 0.9938302040100098

Model prediction: 0.9744548797607422

Model prediction: 0.9985994100570679

Model prediction: 0.9933480024337769

Model prediction: 0.9977060556411743

Model prediction: 0.9938302040100098

Model prediction: 0.9744548797607422

Model prediction: 0.9985994100570679

Model prediction: 0.9933480024337769

Model prediction: 0.9977060556411743

Model prediction: 0.9938302040100098

Model prediction: 0.9744548797607422

Model prediction: 0.9985994100570679

Model prediction: 0.9933480024337769

Model prediction: 0.9977060556411743

Model prediction: 0.9938302040100098

Model prediction: 0.9744548797607422

Model prediction: 0.9985994100570679

Model prediction: 0.9933480024337769

Model prediction: 0.9977060556411743

Model prediction: 0.9938302040100098