A quick reminder:

When your model makes good predictions on the same data that was used to train it but shows poor results with data that hasn& #39;t seen before, we say that the model is overfitting.

The model in the picture is overfitting.

https://abs.twimg.com/emoji/v2/... draggable="false" alt="👇" title="Rückhand Zeigefinger nach unten" aria-label="Emoji: Rückhand Zeigefinger nach unten">
https://abs.twimg.com/emoji/v2/... draggable="false" alt="1⃣" title="Tastenkappe Ziffer 1" aria-label="Emoji: Tastenkappe Ziffer 1"> Train your model on more data

The more data you feed the model, the more likely it will start generalizing (instead of memorizing the training set.)

Look at the relationship between dataset size and error.

(Unfortunately, sometimes there& #39;s no more data.)

https://abs.twimg.com/emoji/v2/... draggable="false" alt="👇" title="Rückhand Zeigefinger nach unten" aria-label="Emoji: Rückhand Zeigefinger nach unten">
https://abs.twimg.com/emoji/v2/... draggable= Train your model on more dataThe more data you feed the model, the more likely it will start generalizing (instead of memorizing the training set.)Look at the relationship between dataset size and error.(Unfortunately, sometimes there& #39;s no more data.)https://abs.twimg.com/emoji/v2/... draggable="false" alt="👇" title="Rückhand Zeigefinger nach unten" aria-label="Emoji: Rückhand Zeigefinger nach unten">" title="https://abs.twimg.com/emoji/v2/... draggable="false" alt="1⃣" title="Tastenkappe Ziffer 1" aria-label="Emoji: Tastenkappe Ziffer 1"> Train your model on more dataThe more data you feed the model, the more likely it will start generalizing (instead of memorizing the training set.)Look at the relationship between dataset size and error.(Unfortunately, sometimes there& #39;s no more data.)https://abs.twimg.com/emoji/v2/... draggable="false" alt="👇" title="Rückhand Zeigefinger nach unten" aria-label="Emoji: Rückhand Zeigefinger nach unten">" class="img-responsive" style="max-width:100%;"/>
https://abs.twimg.com/emoji/v2/... draggable="false" alt="2⃣" title="Tastenkappe Ziffer 2" aria-label="Emoji: Tastenkappe Ziffer 2"> Augment your dataset

You can automatically augment your dataset by transforming existing images in different ways to make the data more diverse.

Some examples:

https://abs.twimg.com/emoji/v2/... draggable="false" alt="▫️" title="Weißes kleines Quadrat" aria-label="Emoji: Weißes kleines Quadrat">Zoom in/out
https://abs.twimg.com/emoji/v2/... draggable="false" alt="▫️" title="Weißes kleines Quadrat" aria-label="Emoji: Weißes kleines Quadrat">Contrast changes
https://abs.twimg.com/emoji/v2/... draggable="false" alt="▫️" title="Weißes kleines Quadrat" aria-label="Emoji: Weißes kleines Quadrat">Horizontal/vertical flips
https://abs.twimg.com/emoji/v2/... draggable="false" alt="▫️" title="Weißes kleines Quadrat" aria-label="Emoji: Weißes kleines Quadrat">Noise addition

https://abs.twimg.com/emoji/v2/... draggable="false" alt="👇" title="Rückhand Zeigefinger nach unten" aria-label="Emoji: Rückhand Zeigefinger nach unten">
https://abs.twimg.com/emoji/v2/... draggable="false" alt="3⃣" title="Tastenkappe Ziffer 3" aria-label="Emoji: Tastenkappe Ziffer 3"> Make your model simpler

You can:

https://abs.twimg.com/emoji/v2/... draggable="false" alt="▫️" title="Weißes kleines Quadrat" aria-label="Emoji: Weißes kleines Quadrat">Reduce the number of layers
https://abs.twimg.com/emoji/v2/... draggable="false" alt="▫️" title="Weißes kleines Quadrat" aria-label="Emoji: Weißes kleines Quadrat">Reduce the number of weights

The more complex your model is, the more capacity it has to memorize the dataset (hence, the easier it will overfit.)

Simplifying the model will force it to generalize.

https://abs.twimg.com/emoji/v2/... draggable="false" alt="👇" title="Rückhand Zeigefinger nach unten" aria-label="Emoji: Rückhand Zeigefinger nach unten">
https://abs.twimg.com/emoji/v2/... draggable= Make your model simplerYou can:https://abs.twimg.com/emoji/v2/... draggable="false" alt="▫️" title="Weißes kleines Quadrat" aria-label="Emoji: Weißes kleines Quadrat">Reduce the number of layershttps://abs.twimg.com/emoji/v2/... draggable="false" alt="▫️" title="Weißes kleines Quadrat" aria-label="Emoji: Weißes kleines Quadrat">Reduce the number of weightsThe more complex your model is, the more capacity it has to memorize the dataset (hence, the easier it will overfit.) Simplifying the model will force it to generalize.https://abs.twimg.com/emoji/v2/... draggable="false" alt="👇" title="Rückhand Zeigefinger nach unten" aria-label="Emoji: Rückhand Zeigefinger nach unten">" title="https://abs.twimg.com/emoji/v2/... draggable="false" alt="3⃣" title="Tastenkappe Ziffer 3" aria-label="Emoji: Tastenkappe Ziffer 3"> Make your model simplerYou can:https://abs.twimg.com/emoji/v2/... draggable="false" alt="▫️" title="Weißes kleines Quadrat" aria-label="Emoji: Weißes kleines Quadrat">Reduce the number of layershttps://abs.twimg.com/emoji/v2/... draggable="false" alt="▫️" title="Weißes kleines Quadrat" aria-label="Emoji: Weißes kleines Quadrat">Reduce the number of weightsThe more complex your model is, the more capacity it has to memorize the dataset (hence, the easier it will overfit.) Simplifying the model will force it to generalize.https://abs.twimg.com/emoji/v2/... draggable="false" alt="👇" title="Rückhand Zeigefinger nach unten" aria-label="Emoji: Rückhand Zeigefinger nach unten">">
https://abs.twimg.com/emoji/v2/... draggable= Make your model simplerYou can:https://abs.twimg.com/emoji/v2/... draggable="false" alt="▫️" title="Weißes kleines Quadrat" aria-label="Emoji: Weißes kleines Quadrat">Reduce the number of layershttps://abs.twimg.com/emoji/v2/... draggable="false" alt="▫️" title="Weißes kleines Quadrat" aria-label="Emoji: Weißes kleines Quadrat">Reduce the number of weightsThe more complex your model is, the more capacity it has to memorize the dataset (hence, the easier it will overfit.) Simplifying the model will force it to generalize.https://abs.twimg.com/emoji/v2/... draggable="false" alt="👇" title="Rückhand Zeigefinger nach unten" aria-label="Emoji: Rückhand Zeigefinger nach unten">" title="https://abs.twimg.com/emoji/v2/... draggable="false" alt="3⃣" title="Tastenkappe Ziffer 3" aria-label="Emoji: Tastenkappe Ziffer 3"> Make your model simplerYou can:https://abs.twimg.com/emoji/v2/... draggable="false" alt="▫️" title="Weißes kleines Quadrat" aria-label="Emoji: Weißes kleines Quadrat">Reduce the number of layershttps://abs.twimg.com/emoji/v2/... draggable="false" alt="▫️" title="Weißes kleines Quadrat" aria-label="Emoji: Weißes kleines Quadrat">Reduce the number of weightsThe more complex your model is, the more capacity it has to memorize the dataset (hence, the easier it will overfit.) Simplifying the model will force it to generalize.https://abs.twimg.com/emoji/v2/... draggable="false" alt="👇" title="Rückhand Zeigefinger nach unten" aria-label="Emoji: Rückhand Zeigefinger nach unten">">
https://abs.twimg.com/emoji/v2/... draggable= Make your model simplerYou can:https://abs.twimg.com/emoji/v2/... draggable="false" alt="▫️" title="Weißes kleines Quadrat" aria-label="Emoji: Weißes kleines Quadrat">Reduce the number of layershttps://abs.twimg.com/emoji/v2/... draggable="false" alt="▫️" title="Weißes kleines Quadrat" aria-label="Emoji: Weißes kleines Quadrat">Reduce the number of weightsThe more complex your model is, the more capacity it has to memorize the dataset (hence, the easier it will overfit.) Simplifying the model will force it to generalize.https://abs.twimg.com/emoji/v2/... draggable="false" alt="👇" title="Rückhand Zeigefinger nach unten" aria-label="Emoji: Rückhand Zeigefinger nach unten">" title="https://abs.twimg.com/emoji/v2/... draggable="false" alt="3⃣" title="Tastenkappe Ziffer 3" aria-label="Emoji: Tastenkappe Ziffer 3"> Make your model simplerYou can:https://abs.twimg.com/emoji/v2/... draggable="false" alt="▫️" title="Weißes kleines Quadrat" aria-label="Emoji: Weißes kleines Quadrat">Reduce the number of layershttps://abs.twimg.com/emoji/v2/... draggable="false" alt="▫️" title="Weißes kleines Quadrat" aria-label="Emoji: Weißes kleines Quadrat">Reduce the number of weightsThe more complex your model is, the more capacity it has to memorize the dataset (hence, the easier it will overfit.) Simplifying the model will force it to generalize.https://abs.twimg.com/emoji/v2/... draggable="false" alt="👇" title="Rückhand Zeigefinger nach unten" aria-label="Emoji: Rückhand Zeigefinger nach unten">">
https://abs.twimg.com/emoji/v2/... draggable="false" alt="4⃣" title="Tastenkappe Ziffer 4" aria-label="Emoji: Tastenkappe Ziffer 4"> Stop the learning process before overfitting

This is known as "Early Stopping."

Identify when overfitting starts happening and stop the learning process before it does.

Plotting the training and validation errors will give you what you need for this.

https://abs.twimg.com/emoji/v2/... draggable="false" alt="👇" title="Rückhand Zeigefinger nach unten" aria-label="Emoji: Rückhand Zeigefinger nach unten">
https://abs.twimg.com/emoji/v2/... draggable= Stop the learning process before overfittingThis is known as "Early Stopping."Identify when overfitting starts happening and stop the learning process before it does. Plotting the training and validation errors will give you what you need for this.https://abs.twimg.com/emoji/v2/... draggable="false" alt="👇" title="Rückhand Zeigefinger nach unten" aria-label="Emoji: Rückhand Zeigefinger nach unten">" title="https://abs.twimg.com/emoji/v2/... draggable="false" alt="4⃣" title="Tastenkappe Ziffer 4" aria-label="Emoji: Tastenkappe Ziffer 4"> Stop the learning process before overfittingThis is known as "Early Stopping."Identify when overfitting starts happening and stop the learning process before it does. Plotting the training and validation errors will give you what you need for this.https://abs.twimg.com/emoji/v2/... draggable="false" alt="👇" title="Rückhand Zeigefinger nach unten" aria-label="Emoji: Rückhand Zeigefinger nach unten">" class="img-responsive" style="max-width:100%;"/>
https://abs.twimg.com/emoji/v2/... draggable="false" alt="5⃣" title="Tastenkappe Ziffer 5" aria-label="Emoji: Tastenkappe Ziffer 5"> Standardize input data

Smaller weights can result in a model less prone to overfit.

Rescaling input data is a way to constraint these weights and keep them from increasing disproportionally.

https://abs.twimg.com/emoji/v2/... draggable="false" alt="👇" title="Rückhand Zeigefinger nach unten" aria-label="Emoji: Rückhand Zeigefinger nach unten">
https://abs.twimg.com/emoji/v2/... draggable="false" alt="7⃣" title="Tastenkappe Ziffer 7" aria-label="Emoji: Tastenkappe Ziffer 7"> L1 and L2 regularization

These refer to a technique that penalizes the loss function to keep the weights of the network constrained.

This means that the network is forced to generalize better because it can& #39;t grow the weights without limit.

https://abs.twimg.com/emoji/v2/... draggable="false" alt="👇" title="Rückhand Zeigefinger nach unten" aria-label="Emoji: Rückhand Zeigefinger nach unten">
Is there anything else you use to prevent your models from overfitting?
You can follow @svpino.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled: