Joerg Marks
2 years ago
1 changed files with 180 additions and 0 deletions
@ -0,0 +1,180 @@ |
|||||
|
{ |
||||
|
"cells": [ |
||||
|
{ |
||||
|
"cell_type": "code", |
||||
|
"execution_count": null, |
||||
|
"id": "8ab45695", |
||||
|
"metadata": {}, |
||||
|
"outputs": [], |
||||
|
"source": [ |
||||
|
"#\n", |
||||
|
"# An example of the minimzer usage in tensor flow\n", |
||||
|
"# the loss function is plotted and the result in terms of a line\n", |
||||
|
"#" |
||||
|
] |
||||
|
}, |
||||
|
{ |
||||
|
"cell_type": "code", |
||||
|
"execution_count": null, |
||||
|
"id": "270932f3", |
||||
|
"metadata": {}, |
||||
|
"outputs": [], |
||||
|
"source": [ |
||||
|
"import numpy as np\n", |
||||
|
"import matplotlib.pyplot as plt\n", |
||||
|
"import tensorflow as tf" |
||||
|
] |
||||
|
}, |
||||
|
{ |
||||
|
"cell_type": "code", |
||||
|
"execution_count": null, |
||||
|
"id": "77cd99a8", |
||||
|
"metadata": {}, |
||||
|
"outputs": [], |
||||
|
"source": [ |
||||
|
"# Define the training data\n", |
||||
|
"train_X = np.asarray([3.3,4.4,5.5,6.71,6.93,4.168,9.779,6.182,7.59,2.167,\n", |
||||
|
" 7.042,10.791,5.313,7.997,5.654,9.27,3.1])\n", |
||||
|
"train_Y = np.asarray([1.7,2.76,2.09,3.19,1.694,1.573,3.366,2.596,2.53,1.221,\n", |
||||
|
" 2.827,3.465,1.65,2.904,2.42,2.94,1.3])\n" |
||||
|
] |
||||
|
}, |
||||
|
{ |
||||
|
"cell_type": "markdown", |
||||
|
"id": "f39cbcd9", |
||||
|
"metadata": {}, |
||||
|
"source": [ |
||||
|
"The input to the model is represented by the train_X \n", |
||||
|
"Y_train represents the target or the truth values for the training data\n", |
||||
|
"The model will recieve train_X and make predictions on the weights\n", |
||||
|
"The difference between these predictions and the actual target values\n", |
||||
|
"train_Y will be used to update the weights and minimize the loss function." |
||||
|
] |
||||
|
}, |
||||
|
{ |
||||
|
"cell_type": "code", |
||||
|
"execution_count": null, |
||||
|
"id": "ed8449c3", |
||||
|
"metadata": {}, |
||||
|
"outputs": [], |
||||
|
"source": [ |
||||
|
"# Define the model to a simple linear regression with only one dense layer and\n", |
||||
|
"# no activation function for the first layer all train_X points are input\n", |
||||
|
"\n", |
||||
|
"# model = tf.keras.models.Sequential([\n", |
||||
|
"# tf.keras.layers.Dense(1, input_shape=[1])\n", |
||||
|
"#])" |
||||
|
] |
||||
|
}, |
||||
|
{ |
||||
|
"cell_type": "code", |
||||
|
"execution_count": null, |
||||
|
"id": "71e072b4", |
||||
|
"metadata": {}, |
||||
|
"outputs": [], |
||||
|
"source": [ |
||||
|
"# This model has 2 dense layers the first with relu activation\n", |
||||
|
"# and the 2nd layer has 1 output unit and uses the default\n", |
||||
|
"# linear activation function.\n", |
||||
|
"\n", |
||||
|
"model = tf.keras.models.Sequential([\n", |
||||
|
" tf.keras.layers.Dense(17, activation='relu',input_shape=[1]),\n", |
||||
|
" tf.keras.layers.Dense(1)\n", |
||||
|
"])" |
||||
|
] |
||||
|
}, |
||||
|
{ |
||||
|
"cell_type": "code", |
||||
|
"execution_count": null, |
||||
|
"id": "5fabf184", |
||||
|
"metadata": {}, |
||||
|
"outputs": [], |
||||
|
"source": [ |
||||
|
"# different optimizer methods can be enabled\n", |
||||
|
"\n", |
||||
|
"model.compile(optimizer=tf.keras.optimizers.Adam(0.01), loss='mean_squared_error')\n", |
||||
|
"#model.compile(optimizer=tf.keras.optimizers.SGD(0.01), loss='mean_squared_error')\n", |
||||
|
"#model.compile(optimizer=tf.keras.optimizers.Adagrad(learning_rate=0.01), loss='mean_squared_error')\n", |
||||
|
"#model.compile(optimizer=tf.keras.optimizers.RMSprop(learning_rate=0.01), loss='mean_squared_error')\n", |
||||
|
"#model.compile(optimizer=tf.keras.optimizers.Ftrl(learning_rate=0.015), loss='mean_squared_error')" |
||||
|
] |
||||
|
}, |
||||
|
{ |
||||
|
"cell_type": "code", |
||||
|
"execution_count": null, |
||||
|
"id": "22c4124f", |
||||
|
"metadata": {}, |
||||
|
"outputs": [], |
||||
|
"source": [ |
||||
|
"# Train the model and access training parameters\n", |
||||
|
"history = model.fit(train_X, train_Y, epochs=60)\n", |
||||
|
"print(history.params)" |
||||
|
] |
||||
|
}, |
||||
|
{ |
||||
|
"cell_type": "code", |
||||
|
"execution_count": null, |
||||
|
"id": "46615960", |
||||
|
"metadata": {}, |
||||
|
"outputs": [], |
||||
|
"source": [ |
||||
|
"# Get the weights of the Dense layer\n", |
||||
|
"weights = model.layers[0].get_weights()\n", |
||||
|
"# Print the weight matrix and bias vector\n", |
||||
|
"print('Weight matrix shape:', weights[0].shape)\n", |
||||
|
"print('Bias vector shape:', weights[1].shape)\n", |
||||
|
"print (weights[0])" |
||||
|
] |
||||
|
}, |
||||
|
{ |
||||
|
"cell_type": "code", |
||||
|
"execution_count": null, |
||||
|
"id": "da12fc5b", |
||||
|
"metadata": {}, |
||||
|
"outputs": [], |
||||
|
"source": [ |
||||
|
"# Plot the loss function\n", |
||||
|
"plt.plot(history.history['loss'])\n", |
||||
|
"plt.title(\"Loss Function\")\n", |
||||
|
"plt.xlabel(\"Epoch\")\n", |
||||
|
"plt.ylabel(\"Loss\")\n", |
||||
|
"plt.show()\n", |
||||
|
"\n", |
||||
|
"# Plot the input data and the predicted values\n", |
||||
|
"plt.plot(train_X, train_Y, 'ro', label=\"Original Data\")\n", |
||||
|
"plt.plot(train_X, model.predict(train_X), label=\"Predicted\")\n", |
||||
|
"plt.legend()\n", |
||||
|
"plt.show()" |
||||
|
] |
||||
|
}, |
||||
|
{ |
||||
|
"cell_type": "code", |
||||
|
"execution_count": null, |
||||
|
"id": "60417d5f", |
||||
|
"metadata": {}, |
||||
|
"outputs": [], |
||||
|
"source": [] |
||||
|
} |
||||
|
], |
||||
|
"metadata": { |
||||
|
"kernelspec": { |
||||
|
"display_name": "Python 3 (ipykernel)", |
||||
|
"language": "python", |
||||
|
"name": "python3" |
||||
|
}, |
||||
|
"language_info": { |
||||
|
"codemirror_mode": { |
||||
|
"name": "ipython", |
||||
|
"version": 3 |
||||
|
}, |
||||
|
"file_extension": ".py", |
||||
|
"mimetype": "text/x-python", |
||||
|
"name": "python", |
||||
|
"nbconvert_exporter": "python", |
||||
|
"pygments_lexer": "ipython3", |
||||
|
"version": "3.8.16" |
||||
|
} |
||||
|
}, |
||||
|
"nbformat": 4, |
||||
|
"nbformat_minor": 5 |
||||
|
} |
Write
Preview
Loading…
Cancel
Save
Reference in new issue