Tensorflow/Keras#

Note

This example requires the tensorflow package to be installed.

Theoretically, tpcp is framework agnostic and can be used with any framework. However, due to the way some frameworks handle their objects, some special handling internally is required. Hence, this example does not only serve as example on how to use tensorflow with tpcp, but also as a test case for these special cases.

When using tpcp with any machine learning framework, you either want to use a pretrained model with a normal pipeline or a train your own model as part of an Optimizable Pipeline. Here we show the second case, as it is more complex, and you are likely able to figure out the first case yourself.

This means, we are planning to perform the following steps:

  1. Create a pipeline that creates and trains a model.

  2. Allow the modification of model hyperparameters.

  3. Run a simple cross-validation to demonstrate the functionality.

This example reimplements the basic MNIST example from the [tensorflow documentation](https://www.tensorflow.org/tutorials/keras/classification).

Some Notes#

In this example we show how to implement a Pipeline that uses tensorflow. You could implement an Algorithm in a similar way. This would actually be easier, as no specific handling of the input data would be required. For a pipeline, we need to create a custom Dataset class, as this is the expected input for a pipeline.

The Dataset#

We are using the normal fashion MNIST dataset for this example It consists of 60.000 images of 28x28 pixels, each with a label. We will ignore the typical train-test split, as we want to do our own cross-validation.

In addition, we will simulate an additional “index level”. In this (and most typical deep learning datasets), each datapoint is one vector for which we can make one prediction. In tpcp, we usually deal with datasets, where you might have multiple pieces of information for each datapoint. For example, one datapoint could be a patient, for which we have an entire time series of measurements. We will simulate this here, by creating the index of our dataset as 1000 groups each containing 60 images.

Other than that, the dataset is pretty standard. Besides the create_index method, we only need to implement the input_as_array and labels_as_array methods that allow us to easily access the data once we selected a single group.

from functools import lru_cache

import numpy as np
import pandas as pd
import tensorflow as tf
from tpcp import Dataset

tf.keras.utils.set_random_seed(812)
tf.config.experimental.enable_op_determinism()


@lru_cache(maxsize=1)
def get_fashion_mnist_data():
    # Note: We throw train and test sets together, as we don't care about the official split here.
    #       We will create our own split later.
    (train_images, train_labels), (test_images, test_labels) = (
        tf.keras.datasets.fashion_mnist.load_data()
    )
    return np.array(list(train_images) + list(test_images)), list(
        train_labels
    ) + list(test_labels)


class FashionMNIST(Dataset):
    def input_as_array(self) -> np.ndarray:
        self.assert_is_single(None, "input_as_array")
        group_id = int(self.group_label.group_id)
        images, _ = get_fashion_mnist_data()
        return (
            images[group_id * 60 : (group_id + 1) * 60].reshape((60, 28, 28))
            / 255
        )

    def labels_as_array(self) -> np.ndarray:
        self.assert_is_single(None, "labels_as_array")
        group_id = int(self.group_label.group_id)
        _, labels = get_fashion_mnist_data()
        return np.array(labels[group_id * 60 : (group_id + 1) * 60])

    def create_index(self) -> pd.DataFrame:
        # There are 60.000 images in total.
        # We simulate 1000 groups of 60 images each.
        return pd.DataFrame({"group_id": list(range(1000))})

We can see our Dataset works as expected:

dataset = FashionMNIST()
dataset[0].input_as_array().shape
Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/train-labels-idx1-ubyte.gz

    0/29515 ━━━━━━━━━━━━━━━━━━━━ 0s 0s/step
29515/29515 ━━━━━━━━━━━━━━━━━━━━ 0s 0us/step
Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/train-images-idx3-ubyte.gz

       0/26421880 ━━━━━━━━━━━━━━━━━━━━ 0s 0s/step
  729088/26421880 ━━━━━━━━━━━━━━━━━━━━ 1s 0us/step
 9379840/26421880 ━━━━━━━━━━━━━━━━━━━━ 0s 0us/step
18325504/26421880 ━━━━━━━━━━━━━━━━━━━━ 0s 0us/step
26421880/26421880 ━━━━━━━━━━━━━━━━━━━━ 0s 0us/step
Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/t10k-labels-idx1-ubyte.gz

   0/5148 ━━━━━━━━━━━━━━━━━━━━ 0s 0s/step
5148/5148 ━━━━━━━━━━━━━━━━━━━━ 0s 0us/step
Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/t10k-images-idx3-ubyte.gz

      0/4422102 ━━━━━━━━━━━━━━━━━━━━ 0s 0s/step
 516096/4422102 ━━━━━━━━━━━━━━━━━━━━ 0s 0us/step
4422102/4422102 ━━━━━━━━━━━━━━━━━━━━ 0s 0us/step

(60, 28, 28)
dataset[0].labels_as_array().shape
(60,)

The Pipeline#

We will create a pipeline that uses a simple neural network to classify the images. In tpcp, all “things” that should be optimized need to be parameters. This means our model itself needs to be a parameter of the pipeline. However, as we don’t have the model yet, as its creation depends on other hyperparameters, we add it as an optional parameter initialized with None. Further, we prefix the parameter name with an underscore, to signify, that this is not a parameter that should be modified manually by the user. This is just convention, and it is up to you to decide how you want to name your parameters.

We further introduce a hyperparameter n_dense_layer_nodes to show how we can influence the model creation.

The optimize method#

To make our pipeline optimizable, it needs to inherit from OptimizablePipeline. Further we need to mark at least one of the parameters as OptiPara using the type annotation. We do this for our _model parameter.

Finally, we need to implement the self_optimize method. This method will get the entire training dataset as input and should update the _model parameter with the trained model. Hence, we first extract the relevant data (remember, each datapoint is 60 images), by concatinating all images over all groups in the dataset. Then we create the Keras model based on the hyperparameters. Finally, we train the model and update the _model parameter.

Here we chose to wrap the method with make_optimize_safe. This decorator will perform some runtime checks to ensure that the method is implemented correctly.

The run method#

The run method expects that the _model parameter is already set (i.e. the pipeline was already optimized). It gets a single datapoint as input (remember, a datapoint is a single group of 60 images). We then extract the data from the datapoint and let the model make a prediction. We store the prediction on our output attribute predictions_. The trailing underscore is a convention to signify, that this is an “result” attribute.

import warnings
from typing import Optional

from tpcp import (
    OptimizablePipeline,
    OptiPara,
    make_action_safe,
    make_optimize_safe,
)
from typing_extensions import Self


class KerasPipeline(OptimizablePipeline):
    n_dense_layer_nodes: int
    n_train_epochs: int
    _model: OptiPara[Optional[tf.keras.Sequential]]

    predictions_: np.ndarray

    def __init__(
        self,
        n_dense_layer_nodes=128,
        n_train_epochs=5,
        _model: Optional[tf.keras.Sequential] = None,
    ):
        self.n_dense_layer_nodes = n_dense_layer_nodes
        self.n_train_epochs = n_train_epochs
        self._model = _model

    @property
    def predicted_labels_(self):
        return np.argmax(self.predictions_, axis=1)

    @make_optimize_safe
    def self_optimize(self, dataset, **_) -> Self:
        data = tf.convert_to_tensor(
            np.vstack([d.input_as_array() for d in dataset])
        )
        labels = tf.convert_to_tensor(
            np.hstack([d.labels_as_array() for d in dataset])
        )

        print(data.shape)
        if self._model is not None:
            warnings.warn("Overwriting existing model!")

        self._model = tf.keras.Sequential(
            [
                tf.keras.layers.Input((28, 28)),
                tf.keras.layers.Flatten(),
                tf.keras.layers.Dense(
                    self.n_dense_layer_nodes, activation="relu"
                ),
                tf.keras.layers.Dense(10),
            ]
        )

        self._model.compile(
            optimizer="adam",
            loss=tf.keras.losses.SparseCategoricalCrossentropy(
                from_logits=True
            ),
            metrics=["accuracy"],
        )

        self._model.fit(data, labels, epochs=self.n_train_epochs)

        return self

    @make_action_safe
    def run(self, datapoint) -> Self:
        if self._model is None:
            raise RuntimeError("Model not trained yet!")
        data = tf.convert_to_tensor(datapoint.input_as_array())

        self.predictions_ = self._model.predict(data)
        return self

Testing the pipeline#

We can now test our pipeline. We will run the optimization using a couple of datapoints (to keep everything fast) and then use run to get the predictions for a single unseen datapoint.

pipeline = KerasPipeline().self_optimize(FashionMNIST()[:10])
p1 = pipeline.run(FashionMNIST()[11])
print(p1.predicted_labels_)
print(FashionMNIST()[11].labels_as_array())
(600, 28, 28)
Epoch 1/5

 1/19 ━━━━━━━━━━━━━━━━━━━━ 9s 521ms/step - accuracy: 0.1250 - loss: 2.3593
19/19 ━━━━━━━━━━━━━━━━━━━━ 1s 2ms/step - accuracy: 0.4933 - loss: 1.5112
Epoch 2/5

 1/19 ━━━━━━━━━━━━━━━━━━━━ 0s 12ms/step - accuracy: 0.7188 - loss: 1.0003
19/19 ━━━━━━━━━━━━━━━━━━━━ 0s 2ms/step - accuracy: 0.7167 - loss: 0.8663
Epoch 3/5

 1/19 ━━━━━━━━━━━━━━━━━━━━ 0s 11ms/step - accuracy: 0.7812 - loss: 0.7427
19/19 ━━━━━━━━━━━━━━━━━━━━ 0s 2ms/step - accuracy: 0.7800 - loss: 0.6830
Epoch 4/5

 1/19 ━━━━━━━━━━━━━━━━━━━━ 0s 12ms/step - accuracy: 0.7812 - loss: 0.6089
19/19 ━━━━━━━━━━━━━━━━━━━━ 0s 2ms/step - accuracy: 0.8133 - loss: 0.5658
Epoch 5/5

 1/19 ━━━━━━━━━━━━━━━━━━━━ 0s 12ms/step - accuracy: 0.7812 - loss: 0.5329
19/19 ━━━━━━━━━━━━━━━━━━━━ 0s 2ms/step - accuracy: 0.8433 - loss: 0.4940

1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 26ms/step
[8 8 6 9 4 0 7 3 7 9 4 8 4 3 7 8 1 4 0 7 9 8 5 5 2 1 3 4 1 9 7 5 9 9 7 8 2
 7 4 7 2 4 7 1 1 7 5 4 8 3 5 9 0 7 4 0 0 9 1 9]
[8 8 0 9 2 0 7 3 7 9 3 8 4 3 7 8 1 4 0 7 9 8 5 5 2 1 3 4 6 7 7 5 9 9 7 8 2
 7 4 7 0 3 5 1 1 5 5 2 8 3 5 9 0 7 3 0 0 7 1 9]

We can see that even with just 5 epochs, the model already performs quite well. To quantify we can calculate the accuracy for this datapoint:

from sklearn.metrics import accuracy_score

accuracy_score(p1.predicted_labels_, FashionMNIST()[11].labels_as_array())
0.8

Cross Validation#

If we want to run a cross validation, we need to formalize the scoring into a function. We will calculate two types of accuracy: First, the accuracy per group and second, the accuracy over all images across all groups. For more information about how this works, check the Custom Scorer example.

from collections.abc import Sequence

from tpcp.validate import Aggregator


class SingleValueAccuracy(Aggregator[tuple[np.ndarray, np.ndarray]]):
    def aggregate(
        self, /, values: Sequence[tuple[np.ndarray, np.ndarray]], **_
    ) -> dict[str, float]:
        return {
            "accuracy": accuracy_score(
                np.hstack([v[0] for v in values]),
                np.hstack([v[1] for v in values]),
            )
        }


single_value_accuracy = SingleValueAccuracy()


def scoring(pipeline, datapoint):
    result: np.ndarray = pipeline.safe_run(datapoint).predicted_labels_
    reference = datapoint.labels_as_array()

    return {
        "accuracy": accuracy_score(result, reference),
        "per_sample": single_value_accuracy((result, reference)),
    }

Now we can run a cross validation. We will only run it on a subset of the data, to keep the runtime manageable.

Note

You might see warnings about retracing of the model. This is because we clone the pipeline before each call to the run method. This is a good idea to ensure that all pipelines are independent of each other, however, might result in some performance overhead.

from tpcp.optimize import Optimize
from tpcp.validate import cross_validate

pipeline = KerasPipeline(n_train_epochs=10)
cv_results = cross_validate(
    Optimize(pipeline), FashionMNIST()[:100], scoring=scoring, cv=3
)
CV Folds:   0%|          | 0/3 [00:00<?, ?it/s](3960, 28, 28)
Epoch 1/10

  1/124 ━━━━━━━━━━━━━━━━━━━━ 56s 462ms/step - accuracy: 0.0938 - loss: 2.5074
 38/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.3846 - loss: 1.6705   
 77/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.4946 - loss: 1.3891
115/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.5500 - loss: 1.2482
124/124 ━━━━━━━━━━━━━━━━━━━━ 1s 1ms/step - accuracy: 0.6919 - loss: 0.8968
Epoch 2/10

  1/124 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step - accuracy: 0.8750 - loss: 0.4848
 40/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.7880 - loss: 0.6149 
 78/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.7944 - loss: 0.5951
119/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.7984 - loss: 0.5871
124/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8111 - loss: 0.5677
Epoch 3/10

  1/124 ━━━━━━━━━━━━━━━━━━━━ 1s 13ms/step - accuracy: 0.8438 - loss: 0.4153
 36/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8176 - loss: 0.5184 
 77/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8265 - loss: 0.5035
118/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8290 - loss: 0.5004
124/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8351 - loss: 0.4936
Epoch 4/10

  1/124 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step - accuracy: 0.8750 - loss: 0.3588
 40/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8353 - loss: 0.4534 
 80/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8399 - loss: 0.4481
121/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8418 - loss: 0.4488
124/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8482 - loss: 0.4475
Epoch 5/10

  1/124 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step - accuracy: 0.8750 - loss: 0.3387
 40/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8501 - loss: 0.4124 
 80/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8533 - loss: 0.4105
119/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8554 - loss: 0.4118
124/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8614 - loss: 0.4122
Epoch 6/10

  1/124 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step - accuracy: 0.8750 - loss: 0.3044
 41/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8623 - loss: 0.3740 
 82/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8629 - loss: 0.3775
124/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8637 - loss: 0.3807
124/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8674 - loss: 0.3840
Epoch 7/10

  1/124 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step - accuracy: 0.9062 - loss: 0.2879
 42/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8647 - loss: 0.3460 
 83/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8673 - loss: 0.3511
124/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8692 - loss: 0.3542
124/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8742 - loss: 0.3577
Epoch 8/10

  1/124 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step - accuracy: 0.8750 - loss: 0.2663
 41/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8719 - loss: 0.3212 
 82/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8760 - loss: 0.3281
123/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8780 - loss: 0.3317
124/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8828 - loss: 0.3357
Epoch 9/10

  1/124 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step - accuracy: 0.9375 - loss: 0.2468
 40/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8890 - loss: 0.2983 
 78/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8880 - loss: 0.3052
117/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8880 - loss: 0.3091
124/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8902 - loss: 0.3147
Epoch 10/10

  1/124 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step - accuracy: 0.9375 - loss: 0.2620
 40/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8983 - loss: 0.2809 
 77/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8960 - loss: 0.2876
118/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8953 - loss: 0.2914
124/124 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8955 - loss: 0.2970


Datapoints:   0%|          | 0/34 [00:00<?, ?it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:   3%|▎         | 1/34 [00:00<00:03,  9.78it/s]WARNING:tensorflow:5 out of the last 5 calls to <function TensorFlowTrainer.make_predict_function.<locals>.one_step_on_data_distributed at 0x7c3166fbaef0> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has reduce_retracing=True option that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.

1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/stepWARNING:tensorflow:6 out of the last 6 calls to <function TensorFlowTrainer.make_predict_function.<locals>.one_step_on_data_distributed at 0x7c3166fbaef0> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has reduce_retracing=True option that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.

2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:   6%|▌         | 2/34 [00:00<00:03,  9.71it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:   9%|▉         | 3/34 [00:00<00:03,  9.80it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  12%|█▏        | 4/34 [00:00<00:03,  9.87it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  15%|█▍        | 5/34 [00:00<00:02,  9.75it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  18%|█▊        | 6/34 [00:00<00:02,  9.79it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 28ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  21%|██        | 7/34 [00:00<00:02,  9.75it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 28ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  24%|██▎       | 8/34 [00:00<00:02,  9.65it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  26%|██▋       | 9/34 [00:00<00:02,  9.70it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 28ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  29%|██▉       | 10/34 [00:01<00:02,  9.72it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  32%|███▏      | 11/34 [00:01<00:02,  9.68it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  35%|███▌      | 12/34 [00:01<00:02,  9.71it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 28ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  38%|███▊      | 13/34 [00:01<00:02,  9.76it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  41%|████      | 14/34 [00:01<00:02,  9.72it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  44%|████▍     | 15/34 [00:01<00:01,  9.74it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  47%|████▋     | 16/34 [00:01<00:01,  9.76it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 28ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  50%|█████     | 17/34 [00:01<00:01,  9.68it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 28ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step


Datapoints:  53%|█████▎    | 18/34 [00:01<00:01,  9.59it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 28ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  56%|█████▌    | 19/34 [00:01<00:01,  9.55it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 28ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  59%|█████▉    | 20/34 [00:02<00:01,  9.45it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  62%|██████▏   | 21/34 [00:02<00:01,  9.60it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  65%|██████▍   | 22/34 [00:02<00:01,  9.71it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  68%|██████▊   | 23/34 [00:02<00:01,  9.70it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  71%|███████   | 24/34 [00:02<00:01,  9.73it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 26ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step

1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  76%|███████▋  | 26/34 [00:02<00:00,  9.79it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  79%|███████▉  | 27/34 [00:02<00:00,  9.84it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  82%|████████▏ | 28/34 [00:02<00:00,  9.88it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 28ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 26ms/step


Datapoints:  85%|████████▌ | 29/34 [00:02<00:00,  9.72it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  88%|████████▊ | 30/34 [00:03<00:00,  9.76it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  91%|█████████ | 31/34 [00:03<00:00,  9.77it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 28ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  94%|█████████▍| 32/34 [00:03<00:00,  9.67it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  97%|█████████▋| 33/34 [00:03<00:00,  9.72it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step

Datapoints: 100%|██████████| 34/34 [00:03<00:00,  9.73it/s]

CV Folds:  33%|███▎      | 1/3 [00:06<00:12,  6.20s/it](4020, 28, 28)
Epoch 1/10

  1/126 ━━━━━━━━━━━━━━━━━━━━ 1:40 808ms/step - accuracy: 0.1250 - loss: 2.3656
 39/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.4280 - loss: 1.6646    
 78/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.5217 - loss: 1.4044
117/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.5722 - loss: 1.2592
126/126 ━━━━━━━━━━━━━━━━━━━━ 1s 1ms/step - accuracy: 0.7000 - loss: 0.8913
Epoch 2/10

  1/126 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step - accuracy: 0.8438 - loss: 0.4830
 39/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.7655 - loss: 0.6245 
 80/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.7718 - loss: 0.6169
120/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.7779 - loss: 0.6067
126/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.7968 - loss: 0.5690
Epoch 3/10

  1/126 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step - accuracy: 0.8750 - loss: 0.3561
 41/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8124 - loss: 0.5224 
 82/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8146 - loss: 0.5205
123/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8174 - loss: 0.5154
126/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8284 - loss: 0.4927
Epoch 4/10

  1/126 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step - accuracy: 0.9375 - loss: 0.3082
 42/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8350 - loss: 0.4816 
 83/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8345 - loss: 0.4809
124/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8364 - loss: 0.4758
126/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8453 - loss: 0.4533
Epoch 5/10

  1/126 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step - accuracy: 0.8750 - loss: 0.3008
 42/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8358 - loss: 0.4395 
 82/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8404 - loss: 0.4361
121/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8440 - loss: 0.4323
126/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8570 - loss: 0.4134
Epoch 6/10

  1/126 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step - accuracy: 0.8750 - loss: 0.2896
 39/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8423 - loss: 0.4071 
 77/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8493 - loss: 0.4046
117/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8537 - loss: 0.4014
126/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8674 - loss: 0.3836
Epoch 7/10

  1/126 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step - accuracy: 0.8750 - loss: 0.2921
 41/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8391 - loss: 0.3811 
 81/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8496 - loss: 0.3775
120/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8552 - loss: 0.3743
126/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8724 - loss: 0.3583
Epoch 8/10

  1/126 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step - accuracy: 0.8750 - loss: 0.2784
 42/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8514 - loss: 0.3570 
 82/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8598 - loss: 0.3542
123/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8650 - loss: 0.3513
126/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8801 - loss: 0.3375
Epoch 9/10

  1/126 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step - accuracy: 0.8750 - loss: 0.2839
 36/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8591 - loss: 0.3376 
 74/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8665 - loss: 0.3345
115/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8720 - loss: 0.3316
126/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8873 - loss: 0.3178
Epoch 10/10

  1/126 ━━━━━━━━━━━━━━━━━━━━ 1s 13ms/step - accuracy: 0.8750 - loss: 0.2753
 41/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8658 - loss: 0.3181 
 82/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8752 - loss: 0.3141
122/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8803 - loss: 0.3116
126/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8940 - loss: 0.3003


Datapoints:   0%|          | 0/33 [00:00<?, ?it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 29ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 26ms/step


Datapoints:   3%|▎         | 1/33 [00:00<00:03,  9.37it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 30ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:   6%|▌         | 2/33 [00:00<00:03,  9.45it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step

1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  12%|█▏        | 4/33 [00:00<00:02,  9.85it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 30ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  15%|█▌        | 5/33 [00:00<00:02,  9.80it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 28ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  18%|█▊        | 6/33 [00:00<00:02,  9.79it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  21%|██        | 7/33 [00:00<00:02,  9.83it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 30ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  24%|██▍       | 8/33 [00:00<00:02,  9.76it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  27%|██▋       | 9/33 [00:00<00:02,  9.82it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  30%|███       | 10/33 [00:01<00:02,  9.87it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 31ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  33%|███▎      | 11/33 [00:01<00:02,  9.72it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 28ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  36%|███▋      | 12/33 [00:01<00:02,  9.67it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  39%|███▉      | 13/33 [00:01<00:02,  9.71it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 32ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 26ms/step


Datapoints:  42%|████▏     | 14/33 [00:01<00:01,  9.61it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  45%|████▌     | 15/33 [00:01<00:01,  9.69it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  48%|████▊     | 16/33 [00:01<00:01,  9.77it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 30ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  52%|█████▏    | 17/33 [00:01<00:01,  9.72it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step

1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  58%|█████▊    | 19/33 [00:01<00:01,  9.84it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 30ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  61%|██████    | 20/33 [00:02<00:01,  9.81it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step

1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  67%|██████▋   | 22/33 [00:02<00:01,  9.87it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 30ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  70%|██████▉   | 23/33 [00:02<00:01,  9.81it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  73%|███████▎  | 24/33 [00:02<00:00,  9.84it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 26ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step

1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 30ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  79%|███████▉  | 26/33 [00:02<00:00,  9.87it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  82%|████████▏ | 27/33 [00:02<00:00,  9.86it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step

1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 28ms/step


Datapoints:  88%|████████▊ | 29/33 [00:02<00:00,  9.88it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step

1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  94%|█████████▍| 31/33 [00:03<00:00,  9.95it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 31ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  97%|█████████▋| 32/33 [00:03<00:00,  9.87it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 28ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints: 100%|██████████| 33/33 [00:03<00:00,  9.86it/s]
Datapoints: 100%|██████████| 33/33 [00:03<00:00,  9.81it/s]

CV Folds:  67%|██████▋   | 2/3 [00:12<00:06,  6.31s/it](4020, 28, 28)
Epoch 1/10

  1/126 ━━━━━━━━━━━━━━━━━━━━ 55s 440ms/step - accuracy: 0.0938 - loss: 2.3129
 39/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.4366 - loss: 1.6292   
 78/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.5247 - loss: 1.3868
119/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.5758 - loss: 1.2432
126/126 ━━━━━━━━━━━━━━━━━━━━ 1s 1ms/step - accuracy: 0.6983 - loss: 0.8965
Epoch 2/10

  1/126 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step - accuracy: 0.8125 - loss: 0.4789
 42/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.7833 - loss: 0.6147 
 82/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.7862 - loss: 0.6108
121/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.7891 - loss: 0.6045
126/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8005 - loss: 0.5778
Epoch 3/10

  1/126 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step - accuracy: 0.8750 - loss: 0.3564
 42/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8241 - loss: 0.5123 
 83/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8221 - loss: 0.5153
124/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8225 - loss: 0.5138
126/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8269 - loss: 0.5017
Epoch 4/10

  1/126 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step - accuracy: 0.9062 - loss: 0.3024
 40/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8485 - loss: 0.4514 
 81/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8448 - loss: 0.4573
123/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8440 - loss: 0.4582
126/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8458 - loss: 0.4515
Epoch 5/10

  1/126 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step - accuracy: 0.9062 - loss: 0.2785
 42/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8627 - loss: 0.4147 
 83/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8595 - loss: 0.4201
124/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8581 - loss: 0.4211
126/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8585 - loss: 0.4165
Epoch 6/10

  1/126 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step - accuracy: 0.9062 - loss: 0.2679
 40/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8762 - loss: 0.3768 
 80/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8729 - loss: 0.3841
120/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8714 - loss: 0.3865
126/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8719 - loss: 0.3833
Epoch 7/10

  1/126 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step - accuracy: 0.9375 - loss: 0.2306
 41/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8849 - loss: 0.3488 
 81/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8824 - loss: 0.3555
121/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8813 - loss: 0.3592
126/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8823 - loss: 0.3594
Epoch 8/10

  1/126 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step - accuracy: 0.9375 - loss: 0.2117
 40/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8915 - loss: 0.3256 
 80/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8886 - loss: 0.3335
121/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8875 - loss: 0.3368
126/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8886 - loss: 0.3366
Epoch 9/10

  1/126 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step - accuracy: 0.9375 - loss: 0.2051
 42/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8954 - loss: 0.3068 
 83/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8935 - loss: 0.3132
124/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8940 - loss: 0.3157
126/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8973 - loss: 0.3152
Epoch 10/10

  1/126 ━━━━━━━━━━━━━━━━━━━━ 1s 12ms/step - accuracy: 0.9375 - loss: 0.1920
 42/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8980 - loss: 0.2892 
 83/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8979 - loss: 0.2962
124/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.8986 - loss: 0.2993
126/126 ━━━━━━━━━━━━━━━━━━━━ 0s 1ms/step - accuracy: 0.9030 - loss: 0.3000


Datapoints:   0%|          | 0/33 [00:00<?, ?it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:   3%|▎         | 1/33 [00:00<00:03,  9.99it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:   6%|▌         | 2/33 [00:00<00:03,  9.93it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 26ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step

1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  12%|█▏        | 4/33 [00:00<00:02, 10.01it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  15%|█▌        | 5/33 [00:00<00:02,  9.83it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step

1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  21%|██        | 7/33 [00:00<00:02,  9.97it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 28ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  24%|██▍       | 8/33 [00:00<00:02,  9.85it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  27%|██▋       | 9/33 [00:00<00:02,  9.86it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  30%|███       | 10/33 [00:01<00:02,  9.87it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  33%|███▎      | 11/33 [00:01<00:02,  9.80it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  36%|███▋      | 12/33 [00:01<00:02,  9.80it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step

1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  42%|████▏     | 14/33 [00:01<00:01,  9.82it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step

1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 26ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  48%|████▊     | 16/33 [00:01<00:01,  9.93it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  52%|█████▏    | 17/33 [00:01<00:01,  9.89it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step

1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  58%|█████▊    | 19/33 [00:01<00:01,  9.91it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  61%|██████    | 20/33 [00:02<00:01,  9.84it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  64%|██████▎   | 21/33 [00:02<00:01,  9.86it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step

1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  70%|██████▉   | 23/33 [00:02<00:01,  9.87it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  73%|███████▎  | 24/33 [00:02<00:00,  9.90it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step

1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  79%|███████▉  | 26/33 [00:02<00:00,  9.89it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step

1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  85%|████████▍ | 28/33 [00:02<00:00,  9.94it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  88%|████████▊ | 29/33 [00:02<00:00,  9.89it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step


Datapoints:  91%|█████████ | 30/33 [00:03<00:00,  9.91it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  94%|█████████▍| 31/33 [00:03<00:00,  9.91it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step


Datapoints:  97%|█████████▋| 32/33 [00:03<00:00,  9.81it/s]
1/2 ━━━━━━━━━━━━━━━━━━━━ 0s 27ms/step
2/2 ━━━━━━━━━━━━━━━━━━━━ 0s 24ms/step

Datapoints: 100%|██████████| 33/33 [00:03<00:00,  9.88it/s]

CV Folds: 100%|██████████| 3/3 [00:18<00:00,  6.16s/it]
CV Folds: 100%|██████████| 3/3 [00:18<00:00,  6.19s/it]

We can now look at the results per group:

cv_results["test__single__accuracy"]
[[0.8333333333333334, 0.8333333333333334, 0.9166666666666666, 0.7666666666666667, 0.8166666666666667, 0.8166666666666667, 0.8, 0.8333333333333334, 0.8333333333333334, 0.85, 0.7666666666666667, 0.8333333333333334, 0.8666666666666667, 0.85, 0.8833333333333333, 0.8666666666666667, 0.8333333333333334, 0.8333333333333334, 0.8, 0.8333333333333334, 0.7833333333333333, 0.7166666666666667, 0.8166666666666667, 0.8333333333333334, 0.8333333333333334, 0.8833333333333333, 0.8833333333333333, 0.85, 0.85, 0.75, 0.75, 0.85, 0.8166666666666667, 0.8833333333333333], [0.8, 0.9, 0.75, 0.8666666666666667, 0.8833333333333333, 0.8166666666666667, 0.8833333333333333, 0.8833333333333333, 0.8333333333333334, 0.8, 0.85, 0.8, 0.8166666666666667, 0.8, 0.7833333333333333, 0.8, 0.8166666666666667, 0.7666666666666667, 0.85, 0.75, 0.8, 0.8333333333333334, 0.7666666666666667, 0.9, 0.8, 0.7833333333333333, 0.8166666666666667, 0.8666666666666667, 0.8833333333333333, 0.9166666666666666, 0.7333333333333333, 0.7833333333333333, 0.9], [0.7833333333333333, 0.9333333333333333, 0.75, 0.8333333333333334, 0.8833333333333333, 0.8666666666666667, 0.8833333333333333, 0.9, 0.9, 0.9, 0.9166666666666666, 0.85, 0.8666666666666667, 0.8833333333333333, 0.85, 0.8, 0.9166666666666666, 0.75, 0.7166666666666667, 0.8666666666666667, 0.8166666666666667, 0.75, 0.7833333333333333, 0.7333333333333333, 0.8666666666666667, 0.8833333333333333, 0.9166666666666666, 0.8, 0.8666666666666667, 0.9, 0.85, 0.85, 0.9]]

Average first per group and then over all groups:

cv_results["test__agg__accuracy"]
array([0.82843137, 0.82525253, 0.84747475])

And the overall accuracy as the average over all samples of all groups within a fold:

cv_results["test__agg__per_sample__accuracy"]
array([0.82843137, 0.82525253, 0.84747475])

Total running time of the script: (0 minutes 30.656 seconds)

Estimated memory usage: 850 MB

Gallery generated by Sphinx-Gallery