mirror of
https://github.com/fenago/data-science.git
synced 2026-05-04 16:41:05 +00:00
246 lines
4.4 KiB
Plaintext
246 lines
4.4 KiB
Plaintext
{
|
|
"cells": [
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"colab": {},
|
|
"colab_type": "code",
|
|
"id": "erRKYnEAzvoF"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"import pandas as pd"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"colab": {},
|
|
"colab_type": "code",
|
|
"id": "jY6Aar2MzzPO"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"file_url = '../Dataset/dataset_44_spambase.csv'"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"colab": {},
|
|
"colab_type": "code",
|
|
"id": "61kW2l6DzzS5"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"df = pd.read_csv(file_url)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"colab": {
|
|
"base_uri": "https://localhost:8080/",
|
|
"height": 255
|
|
},
|
|
"colab_type": "code",
|
|
"id": "VFjr_puO5GBs",
|
|
"outputId": "7b7ae28c-8e12-4209-8bb6-c3d6fe079280"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"print(df)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"colab": {},
|
|
"colab_type": "code",
|
|
"id": "2UG8KAbF7wZD"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"target = df.pop('class')"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"colab": {
|
|
"base_uri": "https://localhost:8080/",
|
|
"height": 221
|
|
},
|
|
"colab_type": "code",
|
|
"id": "sl-jGqPF71Lc",
|
|
"outputId": "ebf9bebb-566b-4866-a4c3-0f75cacd4252"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"print(target)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"colab": {},
|
|
"colab_type": "code",
|
|
"id": "FMhAVQp676Pp"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"from sklearn.ensemble import RandomForestClassifier"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"colab": {},
|
|
"colab_type": "code",
|
|
"id": "PC1vgs2a8DWi"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"seed = 168"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"colab": {},
|
|
"colab_type": "code",
|
|
"id": "M_NSuDKU8BQq"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"rf_model = RandomForestClassifier(random_state=seed)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"colab": {
|
|
"base_uri": "https://localhost:8080/",
|
|
"height": 170
|
|
},
|
|
"colab_type": "code",
|
|
"id": "H4p4BC6D8GD1",
|
|
"outputId": "5ae1448c-a6fc-4033-9e09-f93970f60fe8"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"rf_model.fit(df, target)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"colab": {},
|
|
"colab_type": "code",
|
|
"id": "4udU7PC68KSM"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"preds = rf_model.predict(df)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"colab": {
|
|
"base_uri": "https://localhost:8080/",
|
|
"height": 34
|
|
},
|
|
"colab_type": "code",
|
|
"id": "Fe6XUUSf_YbP",
|
|
"outputId": "09da052f-600d-4b4f-9446-15ff85aaa4ab"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"print(preds)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"colab": {},
|
|
"colab_type": "code",
|
|
"id": "7UtNAj-d-1JC"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"from sklearn.metrics import accuracy_score"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"colab": {},
|
|
"colab_type": "code",
|
|
"id": "Pi0wiBZ9_j6y"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"acc_score = accuracy_score(target, preds)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {
|
|
"colab": {
|
|
"base_uri": "https://localhost:8080/",
|
|
"height": 34
|
|
},
|
|
"colab_type": "code",
|
|
"id": "NxfBXpEB-4Bw",
|
|
"outputId": "4d174d14-1a89-412b-82d4-92aa6ae6d0e3"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"print(acc_score)"
|
|
]
|
|
}
|
|
],
|
|
"metadata": {
|
|
"colab": {
|
|
"collapsed_sections": [],
|
|
"name": "Activity1_1.ipynb",
|
|
"provenance": []
|
|
},
|
|
"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": "3.8.6"
|
|
}
|
|
},
|
|
"nbformat": 4,
|
|
"nbformat_minor": 1
|
|
}
|