manually merge different versions
parent
c285f596b5
commit
c417f3799f
|
@ -11,7 +11,7 @@
|
||||||
},
|
},
|
||||||
{
|
{
|
||||||
"cell_type": "code",
|
"cell_type": "code",
|
||||||
"execution_count": 1,
|
"execution_count": 2,
|
||||||
"id": "3eb339fa-ef85-4544-9ad0-bc22d4de9f1a",
|
"id": "3eb339fa-ef85-4544-9ad0-bc22d4de9f1a",
|
||||||
"metadata": {},
|
"metadata": {},
|
||||||
"outputs": [
|
"outputs": [
|
||||||
|
@ -158,7 +158,7 @@
|
||||||
"4 0.0 3.0 0 "
|
"4 0.0 3.0 0 "
|
||||||
]
|
]
|
||||||
},
|
},
|
||||||
"execution_count": 1,
|
"execution_count": 2,
|
||||||
"metadata": {},
|
"metadata": {},
|
||||||
"output_type": "execute_result"
|
"output_type": "execute_result"
|
||||||
}
|
}
|
||||||
|
@ -190,7 +190,7 @@
|
||||||
},
|
},
|
||||||
{
|
{
|
||||||
"cell_type": "code",
|
"cell_type": "code",
|
||||||
"execution_count": 2,
|
"execution_count": 4,
|
||||||
"id": "6f7e6a3a-63cb-40e2-8746-937c24b184ef",
|
"id": "6f7e6a3a-63cb-40e2-8746-937c24b184ef",
|
||||||
"metadata": {},
|
"metadata": {},
|
||||||
"outputs": [
|
"outputs": [
|
||||||
|
@ -214,7 +214,7 @@
|
||||||
"dtype: int64"
|
"dtype: int64"
|
||||||
]
|
]
|
||||||
},
|
},
|
||||||
"execution_count": 2,
|
"execution_count": 4,
|
||||||
"metadata": {},
|
"metadata": {},
|
||||||
"output_type": "execute_result"
|
"output_type": "execute_result"
|
||||||
}
|
}
|
||||||
|
@ -225,7 +225,7 @@
|
||||||
},
|
},
|
||||||
{
|
{
|
||||||
"cell_type": "code",
|
"cell_type": "code",
|
||||||
"execution_count": 3,
|
"execution_count": 5,
|
||||||
"id": "d1639e92-d401-49fb-a1f1-67250ffa2c81",
|
"id": "d1639e92-d401-49fb-a1f1-67250ffa2c81",
|
||||||
"metadata": {},
|
"metadata": {},
|
||||||
"outputs": [],
|
"outputs": [],
|
||||||
|
@ -233,16 +233,45 @@
|
||||||
"df.dropna(inplace=True)"
|
"df.dropna(inplace=True)"
|
||||||
]
|
]
|
||||||
},
|
},
|
||||||
|
{
|
||||||
|
"cell_type": "markdown",
|
||||||
|
"id": "192da26d-0fb1-4b06-a046-a41b66576ed0",
|
||||||
|
"metadata": {},
|
||||||
|
"source": [
|
||||||
|
"<h1>Preprocessing</h1>\n",
|
||||||
|
"<p>Split</p>"
|
||||||
|
]
|
||||||
|
},
|
||||||
{
|
{
|
||||||
"cell_type": "code",
|
"cell_type": "code",
|
||||||
"execution_count": 4,
|
"execution_count": 17,
|
||||||
|
"id": "24675f41-d48f-4e27-a3d8-e303556ee7d1",
|
||||||
|
"metadata": {},
|
||||||
|
"outputs": [],
|
||||||
|
"source": [
|
||||||
|
"df['goal'] = df['goal'].replace({0: 0, 1: 1, 2: 1, 3: 1, 4: 1})"
|
||||||
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"cell_type": "code",
|
||||||
|
"execution_count": 23,
|
||||||
"id": "d7bf2c46-7885-4dfe-a4e7-8b8439cf0434",
|
"id": "d7bf2c46-7885-4dfe-a4e7-8b8439cf0434",
|
||||||
"metadata": {},
|
"metadata": {},
|
||||||
"outputs": [],
|
"outputs": [],
|
||||||
"source": [
|
"source": [
|
||||||
"# save 'cleaned' dataset as csv file for further processing\n",
|
"# save 'cleaned' dataset as csv file for further processing\n",
|
||||||
|
"import os\n",
|
||||||
|
"os.makedirs('data', exist_ok=True)\n",
|
||||||
"df.to_csv('./data/dataset_cleaned.csv', index=False)"
|
"df.to_csv('./data/dataset_cleaned.csv', index=False)"
|
||||||
]
|
]
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"cell_type": "code",
|
||||||
|
"execution_count": null,
|
||||||
|
"id": "ce134731-e15e-4a2a-aa4e-a28de45fd3af",
|
||||||
|
"metadata": {},
|
||||||
|
"outputs": [],
|
||||||
|
"source": []
|
||||||
}
|
}
|
||||||
],
|
],
|
||||||
"metadata": {
|
"metadata": {
|
||||||
|
@ -261,7 +290,7 @@
|
||||||
"name": "python",
|
"name": "python",
|
||||||
"nbconvert_exporter": "python",
|
"nbconvert_exporter": "python",
|
||||||
"pygments_lexer": "ipython3",
|
"pygments_lexer": "ipython3",
|
||||||
"version": "3.11.7"
|
"version": "3.11.5"
|
||||||
}
|
}
|
||||||
},
|
},
|
||||||
"nbformat": 4,
|
"nbformat": 4,
|
||||||
|
|
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
Loading…
Reference in New Issue