{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
"import matplotlib.pyplot as plt\n",
"import pandas as pd\n",
"import seaborn as sns\n",
"%matplotlib inline"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We want to read in the CSV file."
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"weather = pd.read_csv(\"KNYC.csv\")"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" date | \n",
" actual_mean_temp | \n",
" actual_min_temp | \n",
" actual_max_temp | \n",
" average_min_temp | \n",
" average_max_temp | \n",
" record_min_temp | \n",
" record_max_temp | \n",
" record_min_temp_year | \n",
" record_max_temp_year | \n",
" actual_precipitation | \n",
" average_precipitation | \n",
" record_precipitation | \n",
"
\n",
" \n",
" \n",
" \n",
" 0 | \n",
" 2014-7-1 | \n",
" 81 | \n",
" 72 | \n",
" 89 | \n",
" 68 | \n",
" 83 | \n",
" 52 | \n",
" 100 | \n",
" 1943 | \n",
" 1901 | \n",
" 0.00 | \n",
" 0.12 | \n",
" 2.17 | \n",
"
\n",
" \n",
" 1 | \n",
" 2014-7-2 | \n",
" 82 | \n",
" 72 | \n",
" 91 | \n",
" 68 | \n",
" 83 | \n",
" 56 | \n",
" 100 | \n",
" 2001 | \n",
" 1966 | \n",
" 0.96 | \n",
" 0.13 | \n",
" 1.79 | \n",
"
\n",
" \n",
" 2 | \n",
" 2014-7-3 | \n",
" 78 | \n",
" 69 | \n",
" 87 | \n",
" 68 | \n",
" 83 | \n",
" 54 | \n",
" 103 | \n",
" 1933 | \n",
" 1966 | \n",
" 1.78 | \n",
" 0.12 | \n",
" 2.80 | \n",
"
\n",
" \n",
" 3 | \n",
" 2014-7-4 | \n",
" 70 | \n",
" 65 | \n",
" 74 | \n",
" 68 | \n",
" 84 | \n",
" 55 | \n",
" 102 | \n",
" 1986 | \n",
" 1949 | \n",
" 0.14 | \n",
" 0.13 | \n",
" 1.76 | \n",
"
\n",
" \n",
" 4 | \n",
" 2014-7-5 | \n",
" 72 | \n",
" 63 | \n",
" 81 | \n",
" 68 | \n",
" 84 | \n",
" 53 | \n",
" 101 | \n",
" 1979 | \n",
" 1999 | \n",
" 0.00 | \n",
" 0.12 | \n",
" 3.07 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" date actual_mean_temp actual_min_temp actual_max_temp \\\n",
"0 2014-7-1 81 72 89 \n",
"1 2014-7-2 82 72 91 \n",
"2 2014-7-3 78 69 87 \n",
"3 2014-7-4 70 65 74 \n",
"4 2014-7-5 72 63 81 \n",
"\n",
" average_min_temp average_max_temp record_min_temp record_max_temp \\\n",
"0 68 83 52 100 \n",
"1 68 83 56 100 \n",
"2 68 83 54 103 \n",
"3 68 84 55 102 \n",
"4 68 84 53 101 \n",
"\n",
" record_min_temp_year record_max_temp_year actual_precipitation \\\n",
"0 1943 1901 0.00 \n",
"1 2001 1966 0.96 \n",
"2 1933 1966 1.78 \n",
"3 1986 1949 0.14 \n",
"4 1979 1999 0.00 \n",
"\n",
" average_precipitation record_precipitation \n",
"0 0.12 2.17 \n",
"1 0.13 1.79 \n",
"2 0.12 2.80 \n",
"3 0.13 1.76 \n",
"4 0.12 3.07 "
]
},
"execution_count": 6,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"weather.head()"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"date object\n",
"actual_mean_temp int64\n",
"actual_min_temp int64\n",
"actual_max_temp int64\n",
"average_min_temp int64\n",
"average_max_temp int64\n",
"record_min_temp int64\n",
"record_max_temp int64\n",
"record_min_temp_year int64\n",
"record_max_temp_year int64\n",
"actual_precipitation float64\n",
"average_precipitation float64\n",
"record_precipitation float64\n",
"dtype: object"
]
},
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"weather.dtypes"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [],
"source": [
"weather[\"date\"] = pd.to_datetime(weather[\"date\"]) "
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"date datetime64[ns]\n",
"actual_mean_temp int64\n",
"actual_min_temp int64\n",
"actual_max_temp int64\n",
"average_min_temp int64\n",
"average_max_temp int64\n",
"record_min_temp int64\n",
"record_max_temp int64\n",
"record_min_temp_year int64\n",
"record_max_temp_year int64\n",
"actual_precipitation float64\n",
"average_precipitation float64\n",
"record_precipitation float64\n",
"dtype: object"
]
},
"execution_count": 9,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"weather.dtypes"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"It looks like we have a year's worth of weather data from July 1, 2014 to June 30, 2015. See if there is any missing data."
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" actual_mean_temp | \n",
" actual_min_temp | \n",
" actual_max_temp | \n",
" average_min_temp | \n",
" average_max_temp | \n",
" record_min_temp | \n",
" record_max_temp | \n",
" record_min_temp_year | \n",
" record_max_temp_year | \n",
" actual_precipitation | \n",
" average_precipitation | \n",
" record_precipitation | \n",
"
\n",
" \n",
" \n",
" \n",
" count | \n",
" 365.000000 | \n",
" 365.000000 | \n",
" 365.000000 | \n",
" 365.000000 | \n",
" 365.000000 | \n",
" 365.000000 | \n",
" 365.000000 | \n",
" 365.000000 | \n",
" 365.000000 | \n",
" 365.000000 | \n",
" 365.000000 | \n",
" 365.000000 | \n",
"
\n",
" \n",
" mean | \n",
" 54.736986 | \n",
" 47.246575 | \n",
" 61.734247 | \n",
" 48.016438 | \n",
" 62.079452 | \n",
" 28.243836 | \n",
" 83.731507 | \n",
" 1924.079452 | \n",
" 1958.923288 | \n",
" 0.126164 | \n",
" 0.136822 | \n",
" 2.386137 | \n",
"
\n",
" \n",
" std | \n",
" 18.679979 | \n",
" 18.277156 | \n",
" 19.446971 | \n",
" 14.749176 | \n",
" 16.068765 | \n",
" 20.729107 | \n",
" 13.351306 | \n",
" 38.000820 | \n",
" 34.824357 | \n",
" 0.325577 | \n",
" 0.015734 | \n",
" 1.045702 | \n",
"
\n",
" \n",
" min | \n",
" 11.000000 | \n",
" 2.000000 | \n",
" 19.000000 | \n",
" 27.000000 | \n",
" 38.000000 | \n",
" -15.000000 | \n",
" 54.000000 | \n",
" 1871.000000 | \n",
" 1876.000000 | \n",
" 0.000000 | \n",
" 0.100000 | \n",
" 0.860000 | \n",
"
\n",
" \n",
" 25% | \n",
" 39.000000 | \n",
" 34.000000 | \n",
" 44.000000 | \n",
" 34.000000 | \n",
" 47.000000 | \n",
" 8.000000 | \n",
" 71.000000 | \n",
" 1888.000000 | \n",
" 1933.000000 | \n",
" 0.000000 | \n",
" 0.130000 | \n",
" 1.690000 | \n",
"
\n",
" \n",
" 50% | \n",
" 58.000000 | \n",
" 50.000000 | \n",
" 65.000000 | \n",
" 48.000000 | \n",
" 63.000000 | \n",
" 31.000000 | \n",
" 87.000000 | \n",
" 1920.000000 | \n",
" 1962.000000 | \n",
" 0.000000 | \n",
" 0.140000 | \n",
" 2.160000 | \n",
"
\n",
" \n",
" 75% | \n",
" 72.000000 | \n",
" 64.000000 | \n",
" 80.000000 | \n",
" 63.000000 | \n",
" 78.000000 | \n",
" 47.000000 | \n",
" 96.000000 | \n",
" 1954.000000 | \n",
" 1990.000000 | \n",
" 0.050000 | \n",
" 0.150000 | \n",
" 2.750000 | \n",
"
\n",
" \n",
" max | \n",
" 85.000000 | \n",
" 77.000000 | \n",
" 92.000000 | \n",
" 69.000000 | \n",
" 84.000000 | \n",
" 59.000000 | \n",
" 106.000000 | \n",
" 2015.000000 | \n",
" 2013.000000 | \n",
" 2.540000 | \n",
" 0.170000 | \n",
" 8.280000 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" actual_mean_temp actual_min_temp actual_max_temp average_min_temp \\\n",
"count 365.000000 365.000000 365.000000 365.000000 \n",
"mean 54.736986 47.246575 61.734247 48.016438 \n",
"std 18.679979 18.277156 19.446971 14.749176 \n",
"min 11.000000 2.000000 19.000000 27.000000 \n",
"25% 39.000000 34.000000 44.000000 34.000000 \n",
"50% 58.000000 50.000000 65.000000 48.000000 \n",
"75% 72.000000 64.000000 80.000000 63.000000 \n",
"max 85.000000 77.000000 92.000000 69.000000 \n",
"\n",
" average_max_temp record_min_temp record_max_temp \\\n",
"count 365.000000 365.000000 365.000000 \n",
"mean 62.079452 28.243836 83.731507 \n",
"std 16.068765 20.729107 13.351306 \n",
"min 38.000000 -15.000000 54.000000 \n",
"25% 47.000000 8.000000 71.000000 \n",
"50% 63.000000 31.000000 87.000000 \n",
"75% 78.000000 47.000000 96.000000 \n",
"max 84.000000 59.000000 106.000000 \n",
"\n",
" record_min_temp_year record_max_temp_year actual_precipitation \\\n",
"count 365.000000 365.000000 365.000000 \n",
"mean 1924.079452 1958.923288 0.126164 \n",
"std 38.000820 34.824357 0.325577 \n",
"min 1871.000000 1876.000000 0.000000 \n",
"25% 1888.000000 1933.000000 0.000000 \n",
"50% 1920.000000 1962.000000 0.000000 \n",
"75% 1954.000000 1990.000000 0.050000 \n",
"max 2015.000000 2013.000000 2.540000 \n",
"\n",
" average_precipitation record_precipitation \n",
"count 365.000000 365.000000 \n",
"mean 0.136822 2.386137 \n",
"std 0.015734 1.045702 \n",
"min 0.100000 0.860000 \n",
"25% 0.130000 1.690000 \n",
"50% 0.140000 2.160000 \n",
"75% 0.150000 2.750000 \n",
"max 0.170000 8.280000 "
]
},
"execution_count": 10,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"weather.describe()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Looking at the `count` row, we see there is no missing data. `desribe()` also gives us the 5-number summary (min, 25 percentile, 50 percentile, 75 percentile, and max) for each numerical column, as well as the mean and standard deviation.\n",
"\n",
"Next, let's make some plots. A line plot of the daily mean temperature."
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": 11,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"