Public
Edited
Apr 1, 2023
1 fork
tf = require("@tensorflow/tfjs")
IRIS.csv
Type Table, then Shift-Enter. Ctrl-space for more options.
numOfFeatures = 4
labelMap = ({
"Iris-setosa": 0,
"Iris-versicolor": 1,
"Iris-virginica": 2
})
batchSize = 8
output = []
output
xs = tf.tidy(() => {
return tf.tensor(
iris.map((obj) => [
obj.sepal_length,
obj.sepal_width,
obj.petal_length,
obj.petal_width
])
);
})
xs.toString()
ys = tf.tidy(() => {
return tf.oneHot(
iris.map((obj) => labelMap[obj.species]),
3
);
})
ys.toString()
labels = ["Iris-setosa", "Iris-versicolor", "Iris-virginica"]
mutable accuracy = ""
mutable loss = ""
model = {
const model = tf.sequential({
layers: [
tf.layers.dense({
inputShape: [numOfFeatures],
units: 6,
activation: "relu"
}),
tf.layers.dense({ units: 3, activation: "softmax" })
]
});
await model.compile({
optimizer: "sgd",
loss: "categoricalCrossentropy",
metrics: ["accuracy"]
});
return model;
}
model
.fit(xs, ys, {
epochs: 50,
batchSize: 8,
shuffle: true
})
.then((info) => {
mutable accuracy = "Final accuracy " + info.history.acc[49];
mutable loss = "Final loss " + info.history.loss[49];
})
accuracy
loss
prediction1 = model.predict(tf.tensor([[6, 3, 4.5, 1.5]])).data()
bytesAllocated = {
return tf.memory();
}
Purpose-built for displays of data
Observable is your go-to platform for exploring data and creating expressive data visualizations. Use reactive JavaScript notebooks for prototyping and a collaborative canvas for visual data exploration and dashboard creation.
