perfect-toast-machine - Regression

ShawnHymel / perfect-toast-machine Public

Target: Cortex-M4F 80MHz Clone this project

Neural Network settings

Training settings

Number of training cycles

Please provide a valid number of training cycles (numeric only)

Use learned optimizer

Learning rate

Please provide a valid number for the learning rate (between 0 and 1)

Training processor

Please provide a valid training processor option

Advanced training settings

Validation set size

Please provide a valid number for the train/validate split (between 0 and 1)

Split train/validation set on metadata key

Batch size

Auto-weight classes

Profile int8 model

Neural network architecture

Save

import tensorflow as tf from tensorflow.keras.models import Sequential from tensorflow.keras.layers import Dense, InputLayer, Dropout, Conv1D, Conv2D, Flatten, Reshape, MaxPooling1D, MaxPooling2D, BatchNormalization, TimeDistributed, DepthwiseConv1D, Reshape from tensorflow.keras.optimizers import Adam EPOCHS = 500 # this controls the batch size, or you can manipulate the tf.data.Dataset objects yourself BATCH_SIZE = 32 train_dataset = train_dataset.batch(BATCH_SIZE, drop_remainder=False) validation_dataset = validation_dataset.batch(BATCH_SIZE, drop_remainder=False) # model architecture model = Sequential() # # Middle layers # model.add(Reshape(target_shape=(20, 9), # input_shape=(180,))) # model.add(DepthwiseConv1D(3, # kernel size # strides=1, # padding='same', # depth_multiplier=1, # data_format='channels_last', # activation='relu')) # # Flatten and DNN for classification # model.add(Flatten()) # model.add(Dropout(0.25)) # model.add(Dense(80, # activation=tf.keras.activations.relu)) # model.add(Dropout(0.25)) model.add(Dense(80, activation='relu')) model.add(Dropout(0.25)) model.add(Dense(40, activation='relu')) model.add(Dropout(0.25)) # Final model (for regression) model.add(Dense(classes, name='y_pred', activation='linear')) # this controls the learning rate opt = Adam(learning_rate=0.005, beta_1=0.9, beta_2=0.999) callbacks.append(BatchLoggerCallback(BATCH_SIZE, train_sample_count, epochs=EPOCHS)) # train the neural network model.compile(loss='mean_squared_error', optimizer=opt) model.fit(train_dataset, epochs=EPOCHS, validation_data=validation_dataset, verbose=2, callbacks=callbacks) # Use this flag to disable per-channel quantization for a model. # This can reduce RAM usage for convolutional models, but may have # an impact on accuracy. disable_per_channel_quantization = False

Input layer (180 features)

Dense layer (80 neurons)

Dropout (rate 0.25)

Dense layer (40 neurons)

Dropout (rate 0.25)

Output layer (1 value)

Model

Model version:

Last training performance (validation set)

Loss

130.43

Data explorer (full training set)

Maximum absolute regression error is 0.1, set thresholds.

On-device performance

Engine:

Inferencing time

2 ms.

Peak RAM usage

1.7K

Flash usage

27.6K

Did you know? You have access to the full set of Keras layers through the Expert view (click on to switch),
or can even bring your own model (in PyTorch, Keras or scikit-learn).

Layer type
Dense Fully connected layer, the simplest form of a neural network layer. Use this for processed data, such as the output of a spectral analysis DSP block.
1D Convolution / pooling Learn features that take spatial information into account along a single dimension. Use this for raw data, or for DSP blocks that output spatial data, such as the MFCC block.
2D Convolution / pooling Learn features that take spatial information into account along two dimensions. Use this for raw data, or for DSP blocks that output spatial data, such as the MFCC block.
Reshape Turn one-dimensional data from a DSP block into multi-dimensional data. Use this as an input to a convolutional layer. Use this for deep learning on raw data, or to process MFCC output.
Flatten Flatten multi-dimensional data into a single dimension. You need to flatten data from a convolutional layer before returning.
Dropout Reduce the risk of a model overfitting your dataset by randomly cutting a fraction of network connections during training. Can be helpful if your model's training performance is better than its validation performance.

Complete architectures

Model	Author
LGBM Random Forest Regressor Professional Enterprise The LightGBM Random Forest Regressor applies an ensemble of decision trees with gradient boosting to provide fast, efficient, and accurate regression over large datasets.	Edge Impulse Inc.
XGBoost Random Forest Regressor Professional Enterprise The XGBoost Random Forest Regressor combines the speed and performance of gradient boosting with the robustness of a random forest to achieve high-accuracy regression on complex datasets.	Edge Impulse Inc.
Random Forest Regressor Professional Enterprise The scikit-learn Random Forest Regressor is an ensemble learning method that uses multiple decision trees to predict continuous values, offering high accuracy and handling complex datasets effectively.	Edge Impulse Inc.
Linear Regression Professional Enterprise The scikit-learn Linear Regressor is a straightforward approach that models the relationship between a dependent variable and one or more independent variables using a linear equation to predict continuous outcomes.	Edge Impulse Inc.
Ridge Regression Professional Enterprise The scikit-learn Ridge regressor applies L2 regularization to linear regression, balancing the fit and complexity of the model to improve prediction accuracy on new data by preventing overfitting.	Edge Impulse Inc.
RidgeCV Regression Professional Enterprise The scikit-learn RidgeCV regressor combines ridge regression with built-in cross-validation to automatically tune the regularization strength, ensuring optimal balance between model complexity and performance on unseen data.	Edge Impulse Inc.

The backbone is the foundational element of a neural network model, responsible for extracting meaningful features from input data, enabling subsequent parts to perform anomaly scoring or classification.

Description	Author
EfficientNet V2B0 Supports both RGB and grayscale.	Edge Impulse
MobileNetV2 0.35 Supports both RGB and grayscale.	Edge Impulse
MobileNetV2 0.50 Supports both RGB and grayscale.	Edge Impulse
MobileNetV2 0.75 Supports both RGB and grayscale.	Edge Impulse
MobileNetV2 1.0 Supports both RGB and grayscale.	Edge Impulse
MobileNetV2 0.1 Supports both RGB and grayscale.	Edge Impulse

The scoring function operates on the features extracted by the backbone, and returns anomaly scores for each segment of the image.

Description	Author	Recommended
LGBM Random Forest Regressor The LightGBM Random Forest Regressor applies an ensemble of decision trees with gradient boosting to provide fast, efficient, and accurate regression over large datasets.	Edge Impulse
XGBoost Random Forest Regressor The XGBoost Random Forest Regressor combines the speed and performance of gradient boosting with the robustness of a random forest to achieve high-accuracy regression on complex datasets.	Edge Impulse
Random Forest Regressor The scikit-learn Random Forest Regressor is an ensemble learning method that uses multiple decision trees to predict continuous values, offering high accuracy and handling complex datasets effectively.	Edge Impulse
Linear Regression The scikit-learn Linear Regressor is a straightforward approach that models the relationship between a dependent variable and one or more independent variables using a linear equation to predict continuous outcomes.	Edge Impulse
Ridge Regression The scikit-learn Ridge regressor applies L2 regularization to linear regression, balancing the fit and complexity of the model to improve prediction accuracy on new data by preventing overfitting.	Edge Impulse
RidgeCV Regression The scikit-learn RidgeCV regressor combines ridge regression with built-in cross-validation to automatically tune the regularization strength, ensuring optimal balance between model complexity and performance on unseen data.	Edge Impulse

These users will be notified over email when this job finishes.

	Name
	ShawnHymel

Neural Network settings

Training settings

Augmentation settings

Advanced training settings

Neural network architecture

Model

Last training performance (validation set)

Loss

Confusion matrix (validation set)

Data explorer (full training set)

Settings

On-device performance

Inferencing time

Peak RAM usage

Flash usage

Last training performance (validation set)

Loss

Confusion matrix (validation set)

Data explorer (full training set)

Settings

On-device performance

Inferencing time

Peak RAM usage

Flash usage