Sequential classification¶

• Sequential classification of a tabular MNIST dataset (Task 3573) using a simple neural network.

In [ ]:

import torch.nn
import torch.optim

import openml_pytorch.config
import openml
import logging
import warnings

# Suppress FutureWarning messages
warnings.simplefilter(action='ignore')

############################################################################
# Enable logging in order to observe the progress while running the example.
openml.config.logger.setLevel(logging.DEBUG)
openml_pytorch.config.logger.setLevel(logging.DEBUG)
############################################################################

import torch.nn import torch.optim import openml_pytorch.config import openml import logging import warnings # Suppress FutureWarning messages warnings.simplefilter(action='ignore') ############################################################################ # Enable logging in order to observe the progress while running the example. openml.config.logger.setLevel(logging.DEBUG) openml_pytorch.config.logger.setLevel(logging.DEBUG) ############################################################################

In [ ]:

from openml_pytorch.trainer import OpenMLTrainerModule
from openml_pytorch.trainer import OpenMLDataModule

from openml_pytorch.trainer import OpenMLTrainerModule from openml_pytorch.trainer import OpenMLDataModule

Define the Model¶

In [ ]:

############################################################################
# Define a sequential network that does the initial image reshaping
# and normalization model.
processing_net = torch.nn.Sequential(
    openml_pytorch.layers.Functional(function=torch.Tensor.reshape,
                                                shape=(-1, 1, 28, 28)),
    torch.nn.BatchNorm2d(num_features=1)
)
############################################################################

############################################################################
# Define a sequential network that does the extracts the features from the
# image.
features_net = torch.nn.Sequential(
    torch.nn.Conv2d(in_channels=1, out_channels=32, kernel_size=5),
    torch.nn.LeakyReLU(),
    torch.nn.MaxPool2d(kernel_size=2),
    torch.nn.Conv2d(in_channels=32, out_channels=64, kernel_size=5),
    torch.nn.LeakyReLU(),
    torch.nn.MaxPool2d(kernel_size=2),
)
############################################################################

############################################################################
# Define a sequential network that flattens the features and compiles the
# results into probabilities for each digit.
results_net = torch.nn.Sequential(
    openml_pytorch.layers.Functional(function=torch.Tensor.reshape,
                                                shape=(-1, 4 * 4 * 64)),
    torch.nn.Linear(in_features=4 * 4 * 64, out_features=256),
    torch.nn.LeakyReLU(),
    torch.nn.Dropout(),
    torch.nn.Linear(in_features=256, out_features=10),
)
############################################################################
# openml.config.apikey = 'key'

############################################################################
# The main network, composed of the above specified networks.
model = torch.nn.Sequential(
    processing_net,
    features_net,
    results_net
)
############################################################################

############################################################################ # Define a sequential network that does the initial image reshaping # and normalization model. processing_net = torch.nn.Sequential( openml_pytorch.layers.Functional(function=torch.Tensor.reshape, shape=(-1, 1, 28, 28)), torch.nn.BatchNorm2d(num_features=1) ) ############################################################################ ############################################################################ # Define a sequential network that does the extracts the features from the # image. features_net = torch.nn.Sequential( torch.nn.Conv2d(in_channels=1, out_channels=32, kernel_size=5), torch.nn.LeakyReLU(), torch.nn.MaxPool2d(kernel_size=2), torch.nn.Conv2d(in_channels=32, out_channels=64, kernel_size=5), torch.nn.LeakyReLU(), torch.nn.MaxPool2d(kernel_size=2), ) ############################################################################ ############################################################################ # Define a sequential network that flattens the features and compiles the # results into probabilities for each digit. results_net = torch.nn.Sequential( openml_pytorch.layers.Functional(function=torch.Tensor.reshape, shape=(-1, 4 * 4 * 64)), torch.nn.Linear(in_features=4 * 4 * 64, out_features=256), torch.nn.LeakyReLU(), torch.nn.Dropout(), torch.nn.Linear(in_features=256, out_features=10), ) ############################################################################ # openml.config.apikey = 'key' ############################################################################ # The main network, composed of the above specified networks. model = torch.nn.Sequential( processing_net, features_net, results_net ) ############################################################################

Configure the Data Module¶

• Make sure the target_col is correctly set.

In [ ]:

data_module = OpenMLDataModule(
    type_of_data="dataframe",
    filename_col="class",
    target_mode="categorical",
)

data_module = OpenMLDataModule( type_of_data="dataframe", filename_col="class", target_mode="categorical", )

Configure the Trainer Module¶

In [ ]:

trainer = OpenMLTrainerModule(
    data_module=data_module,
    verbose = True,
    epoch_count = 1,
    callbacks=[],
)
openml_pytorch.config.trainer = trainer

trainer = OpenMLTrainerModule( data_module=data_module, verbose = True, epoch_count = 1, callbacks=[], ) openml_pytorch.config.trainer = trainer

Download the task¶

In [ ]:

# Download the OpenML task for the mnist 784 dataset.
task = openml.tasks.get_task(3573)

# Download the OpenML task for the mnist 784 dataset. task = openml.tasks.get_task(3573)

Run the model on the task¶

In [ ]:

run = openml.runs.run_model_on_task(model, task, avoid_duplicate_runs=False)

run = openml.runs.run_model_on_task(model, task, avoid_duplicate_runs=False)

View loss¶

In [ ]:

trainer.runner.cbs[1].plot_loss()

trainer.runner.cbs[1].plot_loss()

View learning rate¶

In [ ]:

trainer.runner.cbs[1].plot_lr()

trainer.runner.cbs[1].plot_lr()

Publish the run to OpenML¶

In [ ]:

run.publish()

run.publish()