Skip to content

Commit

Permalink
Tema01 colab
Browse files Browse the repository at this point in the history
  • Loading branch information
@johnnynunez
johnnynunez committed Sep 13, 2020
1 parent 6811494 commit f09c239
Show file tree
Hide file tree
Showing 9 changed files with 3,068 additions and 0 deletions.
Binary file added .DS_Store
View file
Binary file not shown.
Binary file added scripts/.DS_Store
View file
Binary file not shown.
Binary file added scripts/tema01/.DS_Store
View file
Binary file not shown.
343 changes: 343 additions & 0 deletions scripts/tema01/01_general_info_colab.ipynb
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,343 @@
{
"nbformat": 4,
"nbformat_minor": 0,
"metadata": {
"colab": {
"name": "01-general-info_colab.ipynb",
"provenance": [],
"collapsed_sections": []
},
"kernelspec": {
"display_name": "Python 3",
"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": "3.8.3"
},
"accelerator": "GPU"
},
"cells": [
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "KdUFcDsdzRyw"
},
"source": [
"# Clonamos el repositorio para obtener los dataSet"
]
},
{
"cell_type": "code",
"metadata": {
"colab_type": "code",
"id": "mHReFf3_y9ms",
"colab": {}
},
"source": [
"!git clone https://github.com/joanby/tensorflow.git"
],
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "vNKZXgtKzU2x"
},
"source": [
"# Damos acceso a nuestro Drive"
]
},
{
"cell_type": "code",
"metadata": {
"colab_type": "code",
"id": "5gu7KWnzzUQ0",
"colab": {}
},
"source": [
"from google.colab import drive\n",
"drive.mount('/content/drive')"
],
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "1gUxIkHWzfHV"
},
"source": [
"# Test it"
]
},
{
"cell_type": "code",
"metadata": {
"colab_type": "code",
"id": "mIQt3jBMzYRE",
"colab": {}
},
"source": [
"!ls '/content/drive/My Drive' "
],
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "mHsK36uN0XB-"
},
"source": [
"# Google colab tools"
]
},
{
"cell_type": "code",
"metadata": {
"colab_type": "code",
"id": "kTzwfUPWzrm4",
"colab": {}
},
"source": [
"from google.colab import files # Para manejar los archivos y, por ejemplo, exportar a su navegador\n",
"import glob # Para manejar los archivos y, por ejemplo, exportar a su navegador\n",
"from google.colab import drive # Montar tu Google drive"
],
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "yQQ8nLiMY6wM",
"colab_type": "text"
},
"source": [
"##Especificando la versión de TensorFlow\n",
"\n",
"Ejecutando \"importar tensorflow\" importará la versión por defecto (actualmente 2.x). Puedes usar la 1.x ejecutando una celda con la \"versión mágica de tensorflow\" **antes de ejecutar \"importar tensorflow\".\n",
"\n",
"### Si no funciona hacer el pip install\n"
]
},
{
"cell_type": "code",
"metadata": {
"id": "1j---G3ZY6wN",
"colab_type": "code",
"colab": {}
},
"source": [
"#!pip install tensorflow==1.14\n",
"%tensorflow_version 1.x"
],
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "S-OIfuWLujbt",
"colab_type": "text"
},
"source": [
"# Importar Tensorflow"
]
},
{
"cell_type": "code",
"metadata": {
"id": "OPSus73fumtP",
"colab_type": "code",
"colab": {}
},
"source": [
"import tensorflow as tf\n",
"print(tf.__version__)"
],
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "Z9ZTNKptt8dA",
"colab_type": "text"
},
"source": [
"# Como funciona TensorFlow"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "IC_M6LlGuAp-",
"colab_type": "text"
},
"source": [
"1. Importación o generación del conjunto de datos.\n",
"2. Transformación y normalización de los datos.\n",
"3. Dividir el conjunto de datos en conjunto de entrenamiento, de validación y de test.\n",
"4. Definir los hiperparámetros del algoritmo"
]
},
{
"cell_type": "code",
"metadata": {
"id": "2-V4YdfJsCL5",
"colab_type": "code",
"colab": {}
},
"source": [
"learning_rate = 0.01\n",
"batch_size = 50\n",
"iterations = 10000"
],
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "7v5UxcIMuINd",
"colab_type": "text"
},
"source": [
"5. Inicializar variables y placeholders"
]
},
{
"cell_type": "code",
"metadata": {
"id": "AbgaGmAFuen3",
"colab_type": "code",
"colab": {}
},
"source": [
"x = tf.constant(30)\n",
"x_input = tf.placeholder(tf.float32, [None, 3]) #None: no sabemos cuantos vectores introducimos Input_size , vectores de 3 coordenadas\n",
"y_input = tf.placeholder(tf.float32, [None, 5]) "
],
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "E6fR1pKJuLaZ",
"colab_type": "text"
},
"source": [
"6. Definir la estructura del modelo del algoritmo.\n",
"$$y = mx+n$$"
]
},
{
"cell_type": "code",
"metadata": {
"id": "t0sPANbqu909",
"colab_type": "code",
"colab": {}
},
"source": [
"y_pred = tf.add(tf.multiply(m_matrix, x_input), n_vector) #add sumar #multiply multiplicar "
],
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "1wrcpp_PuSEL",
"colab_type": "text"
},
"source": [
"7. Declarar la función de pérdidas (loss function)\n",
"$$MSE = \\frac{\\sum_{i=1}^n(y_{actual,i}-y_{pred,i})^2}{n}$$"
]
},
{
"cell_type": "code",
"metadata": {
"id": "PMRrdFr8vAA4",
"colab_type": "code",
"colab": {}
},
"source": [
"loss = tf.reduce_mean(tf.square(y_actual - y_pred)) #reducir la media de los cuadrados"
],
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "FF8ovbE4uUHB",
"colab_type": "text"
},
"source": [
"8. Inicializar y entrenar el modelo anterior. "
]
},
{
"cell_type": "code",
"metadata": {
"id": "8sv0a7czvClk",
"colab_type": "code",
"colab": {}
},
"source": [
"with tf.Session(graph = graph) as session:\n",
" ...\n",
" session.run(...)\n",
" ...."
],
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "wgpFhDMpvEkn",
"colab_type": "code",
"colab": {}
},
"source": [
"session = tf.Session(graph = graph)\n",
"...\n",
"session.run(...)\n",
"..."
],
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "rFYN2Z4yuV79",
"colab_type": "text"
},
"source": [
"9. Evaluación del modelo \n",
"10. Ajustar los hiper parámetros\n",
"11. Publicar (subir a producción) y predecir nuevos resultados"
]
}
]
}
Loading

0 comments on commit f09c239

Please sign in to comment.