markdown stringlengths 0 1.02M | code stringlengths 0 832k | output stringlengths 0 1.02M | license stringlengths 3 36 | path stringlengths 6 265 | repo_name stringlengths 6 127 |
|---|---|---|---|---|---|
We provide `distance` and `radial_velocity` to prepare the data for reflex correction, which we explain below. | type(skycoord) | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
The result is an Astropy `SkyCoord` object ([documentation here](https://docs.astropy.org/en/stable/api/astropy.coordinates.SkyCoord.htmlastropy.coordinates.SkyCoord)), which provides `transform_to`, so we can transform the coordinates to other frames. | import gala.coordinates as gc
transformed = skycoord.transform_to(gc.GD1Koposov10)
type(transformed) | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
The result is another `SkyCoord` object, now in the `GD1Koposov10` frame. The next step is to correct the proper motion measurements from Gaia for reflex due to the motion of our solar system around the Galactic center.When we created `skycoord`, we provided `distance` and `radial_velocity` as arguments, which means we... | gd1_coord = gc.reflex_correct(transformed)
type(gd1_coord) | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
The result is a `SkyCoord` object that contains * The transformed coordinates as attributes named `phi1` and `phi2`, which represent right ascension and declination in the `GD1Koposov10` frame.* The transformed and corrected proper motions as `pm_phi1_cosphi2` and `pm_phi2`.We can select the coordinates like this: | phi1 = gd1_coord.phi1
phi2 = gd1_coord.phi2 | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
And plot them like this: | plt.plot(phi1, phi2, 'ko', markersize=0.1, alpha=0.2)
plt.xlabel('ra (degree GD1)')
plt.ylabel('dec (degree GD1)'); | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
Remember that we started with a rectangle in GD-1 coordinates. When transformed to ICRS, it's a non-rectangular polygon. Now that we have transformed back to GD-1 coordinates, it's a rectangle again. Pandas DataFrameAt this point we have three objects containing different subsets of the data. | type(results)
type(gaia_data)
type(gd1_coord) | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
On one hand, this makes sense, since each object provides different capabilities. But working with three different object types can be awkward.It will be more convenient to choose one object and get all of the data into it. We'll use a Pandas DataFrame, for two reasons:1. It provides capabilities that are pretty much... | import pandas as pd
df = results.to_pandas()
df.shape | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
`DataFrame` provides `shape`, which shows the number of rows and columns.It also provides `head`, which displays the first few rows. It is useful for spot-checking large results as you go along. | df.head() | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
Python detail: `shape` is an attribute, so we can display it's value without calling it as a function; `head` is a function, so we need the parentheses. Now we can extract the columns we want from `gd1_coord` and add them as columns in the `DataFrame`. `phi1` and `phi2` contain the transformed coordinates. | df['phi1'] = gd1_coord.phi1
df['phi2'] = gd1_coord.phi2
df.shape | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
`pm_phi1_cosphi2` and `pm_phi2` contain the components of proper motion in the transformed frame. | df['pm_phi1'] = gd1_coord.pm_phi1_cosphi2
df['pm_phi2'] = gd1_coord.pm_phi2
df.shape | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
**Detail:** If you notice that `SkyCoord` has an attribute called `proper_motion`, you might wonder why we are not using it.We could have: `proper_motion` contains the same data as `pm_phi1_cosphi2` and `pm_phi2`, but in a different format. Plot proper motionNow we are ready to replicate one of the panels in Figure 1 ... | phi2 = df['phi2']
type(phi2) | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
The result is a `Series`, which is the structure Pandas uses to represent columns.We can use a comparison operator, `>`, to compare the values in a `Series` to a constant. | phi2_min = -1.0 * u.deg
phi2_max = 1.0 * u.deg
mask = (df['phi2'] > phi2_min)
type(mask)
mask.dtype | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
The result is a `Series` of Boolean values, that is, `True` and `False`. | mask.head() | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
A Boolean `Series` is sometimes called a "mask" because we can use it to mask out some of the rows in a `DataFrame` and select the rest, like this: | subset = df[mask]
type(subset) | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
`subset` is a `DataFrame` that contains only the rows from `df` that correspond to `True` values in `mask`.The previous mask selects all stars where `phi2` exceeds `phi2_min`; now we'll select stars where `phi2` falls between `phi2_min` and `phi2_max`. | phi_mask = ((df['phi2'] > phi2_min) &
(df['phi2'] < phi2_max)) | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
The `&` operator computes "logical AND", which means the result is true where elements from both Boolean `Series` are true.The sum of a Boolean `Series` is the number of `True` values, so we can use `sum` to see how many stars are in the selected region. | phi_mask.sum() | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
And we can use `phi1_mask` to select stars near the centerline, which are more likely to be in GD-1. | centerline = df[phi_mask]
len(centerline) | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
Here's a scatter plot of proper motion for the selected stars. | pm1 = centerline['pm_phi1']
pm2 = centerline['pm_phi2']
plt.plot(pm1, pm2, 'ko', markersize=0.1, alpha=0.1)
plt.xlabel('Proper motion phi1 (GD1 frame)')
plt.ylabel('Proper motion phi2 (GD1 frame)'); | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
Looking at these results, we see a large cluster around (0, 0), and a smaller cluster near (0, -10).We can use `xlim` and `ylim` to set the limits on the axes and zoom in on the region near (0, 0). | pm1 = centerline['pm_phi1']
pm2 = centerline['pm_phi2']
plt.plot(pm1, pm2, 'ko', markersize=0.3, alpha=0.3)
plt.xlabel('Proper motion phi1 (GD1 frame)')
plt.ylabel('Proper motion phi2 (GD1 frame)')
plt.xlim(-12, 8)
plt.ylim(-10, 10); | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
Now we can see the smaller cluster more clearly.You might notice that our figure is less dense than the one in the paper. That's because we started with a set of stars from a relatively small region. The figure in the paper is based on a region about 10 times bigger.In the next lesson we'll go back and select stars f... | pm1_min = -8.9
pm1_max = -6.9
pm2_min = -2.2
pm2_max = 1.0 | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
To draw these bounds, we'll make two lists containing the coordinates of the corners of the rectangle. | pm1_rect = [pm1_min, pm1_min, pm1_max, pm1_max, pm1_min] * u.mas/u.yr
pm2_rect = [pm2_min, pm2_max, pm2_max, pm2_min, pm2_min] * u.mas/u.yr | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
Here's what the plot looks like with the bounds we chose. | plt.plot(pm1, pm2, 'ko', markersize=0.3, alpha=0.3)
plt.plot(pm1_rect, pm2_rect, '-')
plt.xlabel('Proper motion phi1 (GD1 frame)')
plt.ylabel('Proper motion phi2 (GD1 frame)')
plt.xlim(-12, 8)
plt.ylim(-10, 10); | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
To select rows that fall within these bounds, we'll use the following function, which uses Pandas operators to make a mask that selects rows where `series` falls between `low` and `high`. | def between(series, low, high):
"""Make a Boolean Series.
series: Pandas Series
low: lower bound
high: upper bound
returns: Boolean Series
"""
return (series > low) & (series < high) | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
The following mask select stars with proper motion in the region we chose. | pm_mask = (between(df['pm_phi1'], pm1_min, pm1_max) &
between(df['pm_phi2'], pm2_min, pm2_max)) | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
Again, the sum of a Boolean series is the number of `True` values. | pm_mask.sum() | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
Now we can use this mask to select rows from `df`. | selected = df[pm_mask]
len(selected) | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
These are the stars we think are likely to be in GD-1. Let's see what they look like, plotting their coordinates (not their proper motion). | phi1 = selected['phi1']
phi2 = selected['phi2']
plt.plot(phi1, phi2, 'ko', markersize=0.5, alpha=0.5)
plt.xlabel('ra (degree GD1)')
plt.ylabel('dec (degree GD1)'); | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
Now that's starting to look like a tidal stream! Saving the DataFrameAt this point we have run a successful query and cleaned up the results; this is a good time to save the data.To save a Pandas `DataFrame`, one option is to convert it to an Astropy `Table`, like this: | selected_table = Table.from_pandas(selected)
type(selected_table) | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
Then we could write the `Table` to a FITS file, as we did in the previous lesson. But Pandas provides functions to write DataFrames in other formats; to see what they are [find the functions here that begin with `to_`](https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.html).One of the best op... | filename = 'gd1_dataframe.hdf5'
df.to_hdf(filename, 'df', mode='w') | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
Because an HDF5 file can contain more than one Dataset, we have to provide a name, or "key", that identifies the Dataset in the file.We could use any string as the key, but in this example I use the variable name `df`. Exercise We're going to need `centerline` and `selected` later as well. Write a line or two of code... | # Solution
centerline.to_hdf(filename, 'centerline')
selected.to_hdf(filename, 'selected') | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
**Detail:** Reading and writing HDF5 tables requires a library called `PyTables` that is not always installed with Pandas. You can install it with pip like this:```pip install tables```If you install it using Conda, the name of the package is `pytables`.```conda install pytables``` We can use `ls` to confirm that the ... | !ls -lh gd1_dataframe.hdf5 | -rw-rw-r-- 1 downey downey 17M Nov 18 19:06 gd1_dataframe.hdf5
| MIT | 03_motion.ipynb | abostroem/AstronomicalData |
If you are using Windows, `ls` might not work; in that case, try:```!dir gd1_dataframe.hdf5```We can read the file back like this: | read_back_df = pd.read_hdf(filename, 'df')
read_back_df.shape | _____no_output_____ | MIT | 03_motion.ipynb | abostroem/AstronomicalData |
Preparation for ColabMake sure you're running a GPU runtime; if not, select "GPU" as the hardware accelerator in Runtime > Change Runtime Type in the menu. The next cells will install the `clip` package and its dependencies, and check if PyTorch 1.7.1 or later is installed. | # ! pip install ftfy regex tqdm
# ! pip install git+https://github.com/openai/CLIP.git
import numpy as np
import torch
import clip
# from tqdm.notebook import tqdm
from tqdm import tqdm
print("Torch version:", torch.__version__)
assert torch.__version__.split(".") >= ["1", "7", "1"], "PyTorch 1.7.1 or later is requir... | Torch version: 1.9.0
| MIT | notebooks/Prompt_Engineering_for_CIFAR.ipynb | KAndHisC/CLIP |
Loading the modelDownload and instantiate a CLIP model using the `clip` module that we just installed. | print(clip.available_models())
print(os.getcwd())
# model, preprocess = clip.load("ViT-B/32")
model, preprocess = clip.load("../models/200m0.988.pt")
input_resolution = model.visual.input_resolution
context_length = model.context_length
vocab_size = model.vocab_size
print("Model parameters:", f"{np.sum([int(np.prod(p.... | Model parameters: 151,277,313
Input resolution: 224
Context length: 77
Vocab size: 49408
| MIT | notebooks/Prompt_Engineering_for_CIFAR.ipynb | KAndHisC/CLIP |
Preparing ImageNet labels and promptsThe following cell contains the 1,000 labels for the ImageNet dataset, followed by the text templates we'll use as "prompt engineering". | classes = [
'apple',
'aquarium fish',
'baby',
'bear',
'beaver',
'bed',
'bee',
'beetle',
'bicycle',
'bottle',
'bowl',
'boy',
'bridge',
'bus',
'butterfly',
'camel',
'can',
'castle',
'caterpillar',
'cattle',
'chair',
'chimpanzee',
... | _____no_output_____ | MIT | notebooks/Prompt_Engineering_for_CIFAR.ipynb | KAndHisC/CLIP |
A subset of these class names are modified from the default ImageNet class names sourced from Anish Athalye's imagenet-simple-labels.These edits were made via trial and error and concentrated on the lowest performing classes according to top_1 and top_5 accuracy on the ImageNet training set for the RN50, RN101, and RN5... |
print(f"{len(classes)} classes, {len(templates)} templates") | 100 classes, 18 templates
| MIT | notebooks/Prompt_Engineering_for_CIFAR.ipynb | KAndHisC/CLIP |
A similar, intuition-guided trial and error based on the ImageNet training set was used for templates. This list is pretty haphazard and was gradually made / expanded over the course of about a year of the project and was revisited / tweaked every few months. A surprising / weird thing was adding templates intended to ... | # ! pip install git+https://github.com/modestyachts/ImageNetV2_pytorch
# from imagenetv2_pytorch import ImageNetV2Dataset
# images = ImageNetV2Dataset(transform=preprocess)
from torchvision.datasets import CIFAR100
import os
cifar100 = CIFAR100(root=os.path.expanduser("./data/cifar100"), download=True, train=True, t... | Files already downloaded and verified
| MIT | notebooks/Prompt_Engineering_for_CIFAR.ipynb | KAndHisC/CLIP |
Creating zero-shot classifier weights | def zeroshot_classifier(classnames, templates):
with torch.no_grad():
zeroshot_weights = []
for classname in tqdm(classnames):
texts = [template.format(classname) for template in templates] #format with class
texts = clip.tokenize(texts).cuda() #tokenize
class_emb... | 100%|██████████| 100/100 [00:01<00:00, 74.94it/s]
| MIT | notebooks/Prompt_Engineering_for_CIFAR.ipynb | KAndHisC/CLIP |
Zero-shot prediction | def accuracy(output, target, topk=(1,)):
pred = output.topk(max(topk), 1, True, True)[1].t()
correct = pred.eq(target.view(1, -1).expand_as(pred))
return [float(correct[:k].reshape(-1).float().sum(0, keepdim=True).cpu().numpy()) for k in topk]
with torch.no_grad():
top1, top5, n = 0., 0., 0.
for i, ... | 100%|██████████| 1563/1563 [00:43<00:00, 35.92it/s] | MIT | notebooks/Prompt_Engineering_for_CIFAR.ipynb | KAndHisC/CLIP |
Car Accidents During COVID Stay-At-Home Periods AnalysisThis notebook is using data collected and cleaned in the `car_accidents_eda.ipynb` notebook, and will be used for examining differences state-by-state for severity and frequency of car accidents. Tableau graphs will be used for final graphs, but this notebook gi... | import pickle
import pandas as pd
import numpy as np | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
Setup First import the dataset created and prepared for analysis > | # open our master dataset for analysis
with open('pickle/car_accidents_master.pickle','rb') as read_file:
car_accidents_master = pickle.load(read_file) | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
Note that we are looking at our dataset before it was narrowed for modeling (e.g. only severity 2 & 3, only using mapquest-sourced data). For the analysis we'll be doing here we can look at the full dataset.But first we'll narrow to only the columns we'll be using, which are date, state, whether in period of shutdown,... | accident_covid = car_accidents_master[['Severity','Date',
'State','Year','Month',
'Day','Day_Of_Week',
'Shut_Down'
]] | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
Now we'll look at amount of accidents in each states > | accident_covid.State.value_counts() | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
And then see if we narrowed the number of states, how many we'd need to get to at least 75% of total accidents > | accident_covid.State.value_counts()[:20].sum()/accident_covid.State.value_counts().sum()
accident_covid.State.value_counts()[:20] | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
First we'll focus on the state with the highest amount of accidents, as they will provided more data to then draw better conclusions from and also are of greater interest since more need to address car accidents. We'll focus on the 20 states that have the most car accidents, which also represent 86% of total accidents... | def severity_percent(state_data):
'''
Takes in state data and returns
a list of percentages for each
severity level to total accidents.
'''
percentages = []
if 1 not in sorted(state_data.Severity.unique()):
percentages.append(0)
for x in sorted(state_data.Severity.unique())... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
- Top 20 States Analysis California First create periods we'll want to compare for the state > | # full state period
ca_ttl = accident_covid[(accident_covid.State == 'CA')]
# shutdown period
ca_sd = accident_covid[
(accident_covid.State == 'CA') &
(accident_covid.Shut_Down == 1)]
# 2019
# note CA is still in shutdown, so we're comparing to when dataset ended in 2020
# not when their stay-at-home ord... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
FrequencyNow compare frequencies (total number) of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018, & 2017 being in same month-day date range as shut-down period) > | # find total frequencies for state total and in each period
ca_freqs = [ca_ttl.shape[0],
ca_sd.shape[0],
ca_2019.shape[0],
ca_2018.shape[0],
ca_2017.shape[0]
]
# then develop a simple dataframe to compare them
ca_freq_data = pd.DataFrame({'Timeframe': ['Tota... | Change Frequency: 76.44%
| AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
SeverityNow compare severities of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018 & 2017 being in same month-day date range as shut-down period) > | # first find percentages of severity ranks to total accidents for
# total and for each period
ca_sp_ttl = severity_percent(ca_ttl)
ca_sp_sd = severity_percent(ca_sd)
ca_sp_2019 = severity_percent(ca_2019)
ca_sp_2018 = severity_percent(ca_2018)
ca_sp_2017 = severity_percent(ca_2017)
# then again create a simple da... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
CA Takeaways:- See increase in fequency during shut down period compared to same periods in past three years- Seeing increase in 'edges' of severity -- 1 & 4 -- of car accidents with lowering of 'middle' severities -- 2 & 3 -- for shutdown period compared to average ttl and past year's prior periods - Texas First crea... | # full state period
tx_ttl = accident_covid[(accident_covid.State == 'TX')]
# shutdown period
tx_sd = accident_covid[
(accident_covid.State == 'TX') &
(accident_covid.Shut_Down == 1)]
# 2019
tx_2019 = accident_covid[(car_accidents_master['Date'] > '2019-03-31') &
(car_accidents_mas... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
FrequencyNow compare frequencies (total number) of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018, & 2017 being in same month-day date range as shut-down period) > | # find total frequencies for state total and in each period
tx_freqs = [tx_ttl.shape[0],
tx_sd.shape[0],
tx_2019.shape[0],
tx_2018.shape[0],
tx_2017.shape[0]
]
# then develop a simple dataframe to compare them
tx_freq_data = pd.DataFrame({'Timeframe': ['Tota... | Change Frequency: -37.65%
| AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
SeverityNow compare severities of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018 & 2017 being in same month-day date range as shut-down period) > | # first find percentages of severity ranks to total accidents for
# total and for each period
tx_sp_ttl = severity_percent(tx_ttl)
tx_sp_sd = severity_percent(tx_sd)
tx_sp_2019 = severity_percent(tx_2019)
tx_sp_2018 = severity_percent(tx_2018)
tx_sp_2017 = severity_percent(tx_2017)
# then again create a simple da... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
TX Takeaways:- Frequency: Unlike CA, TX had a drop in frequency during shutdown compared to past periods- Severity: Similar to CA, still the 'edges' of severity (1 & 4) growing, 4 in particular to a greater degree than CA - FloridaFirst create periods we'll want to compare for the state > | # full state period
fl_ttl = accident_covid[(accident_covid.State == 'FL')]
# shutdown period
fl_sd = accident_covid[
(accident_covid.State == 'FL') &
(accident_covid.Shut_Down == 1)]
# 2019
fl_2019 = accident_covid[(car_accidents_master['Date'] > '2019-04-03') &
(car_accidents_mas... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
FrequencyNow compare frequencies (total number) of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018, & 2017 being in same month-day date range as shut-down period) > | # find total frequencies for state total and in each period
fl_freqs = [fl_ttl.shape[0],
fl_sd.shape[0],
fl_2019.shape[0],
fl_2018.shape[0],
fl_2017.shape[0]
]
# then develop a simple dataframe to compare them
fl_freq_data = pd.DataFrame({'Timeframe': ['Tota... | Change Frequency: 6.62%
| AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
SeverityNow compare severities of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018 & 2017 being in same month-day date range as shut-down period) > | # first find percentages of severity ranks to total accidents for
# total and for each period
fl_sp_ttl = severity_percent(fl_ttl)
fl_sp_sd = severity_percent(fl_sd)
fl_sp_2019 = severity_percent(fl_2019)
fl_sp_2018 = severity_percent(fl_2018)
fl_sp_2017 = severity_percent(fl_2017)
# then again create a simple da... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
FL Takeaways:- Frequency: Not really much change positively or negatively for shutdown compared to past periods- Severity: Don't see a lot of difference in severity 4 as with CA & TX, but do see significant increase in 1s while 2 & 3 dropped - South CarolinaFirst create periods we'll want to compare for the state > | # full state period
sc_ttl = accident_covid[(accident_covid.State == 'SC')]
# shutdown period
sc_sd = accident_covid[
(accident_covid.State == 'SC') &
(accident_covid.Shut_Down == 1)]
# 2019
sc_2019 = accident_covid[(car_accidents_master['Date'] > '2019-04-06') &
(car_accidents_mas... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
FrequencyNow compare frequencies (total number) of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018, & 2017 being in same month-day date range as shut-down period) > | # find total frequencies for state total and in each period
sc_freqs = [sc_ttl.shape[0],
sc_sd.shape[0],
sc_2019.shape[0],
sc_2018.shape[0],
sc_2017.shape[0]
]
# then develop a simple dataframe to compare them
sc_freq_data = pd.DataFrame({'Timeframe': ['Tota... | Change Frequency: 23.58%
| AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
SeverityNow compare severities of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018 & 2017 being in same month-day date range as shut-down period) > | # first find percentages of severity ranks to total accidents for
# total and for each period
sc_sp_ttl = severity_percent(sc_ttl)
sc_sp_sd = severity_percent(sc_sd)
sc_sp_2019 = severity_percent(sc_2019)
sc_sp_2018 = severity_percent(sc_2018)
sc_sp_2017 = severity_percent(sc_2017)
# then again create a simple da... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
SC Takeaways:- Frequency: no significant difference for shutdown period. 2017 seems low, but this may be due to less data collected that year. - Severity: in general, seemed to lower in severity. See more 1 & 2 and less 3 & 4 - North CarolinaFirst create periods we'll want to compare for the state > | # full state period
nc_ttl = accident_covid[(accident_covid.State == 'NC')]
# shutdown period
nc_sd = accident_covid[
(accident_covid.State == 'NC') &
(accident_covid.Shut_Down == 1)]
# 2019
nc_2019 = accident_covid[(car_accidents_master['Date'] > '2019-03-30') &
(car_accidents_mas... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
FrequencyNow compare frequencies (total number) of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018, & 2017 being in same month-day date range as shut-down period) > | # find total frequencies for state total and in each period
nc_freqs = [nc_ttl.shape[0],
nc_sd.shape[0],
nc_2019.shape[0],
nc_2018.shape[0],
nc_2017.shape[0]
]
# then develop a simple dataframe to compare them
nc_freq_data = pd.DataFrame({'Timeframe': ['Tota... | Change Frequency: 22.40%
| AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
SeverityNow compare severities of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018 & 2017 being in same month-day date range as shut-down period) > | # first find percentages of severity ranks to total accidents for
# total and for each period
nc_sp_ttl = severity_percent(nc_ttl)
nc_sp_sd = severity_percent(nc_sd)
nc_sp_2019 = severity_percent(nc_2019)
nc_sp_2018 = severity_percent(nc_2018)
nc_sp_2017 = severity_percent(nc_2017)
# then again create a simple da... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
NC Takeaways:- Fequency: No significant change during shutdown period- Severity: Do see an increase in level 4 severity, but greatest change is increase in severity 1 with drop in severity 2 - New YorkFirst create periods we'll want to compare for the state > | # full state period
ny_ttl = accident_covid[(accident_covid.State == 'NY')]
# shutdown period
ny_sd = accident_covid[
(accident_covid.State == 'NY') &
(accident_covid.Shut_Down == 1)]
# 2019
ny_2019 = accident_covid[(car_accidents_master['Date'] > '2019-03-22') &
(car_accidents_mas... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
FrequencyNow compare frequencies (total number) of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018, & 2017 being in same month-day date range as shut-down period) > | # find total frequencies for state total and in each period
ny_freqs = [ny_ttl.shape[0],
ny_sd.shape[0],
ny_2019.shape[0],
ny_2018.shape[0],
ny_2017.shape[0]
]
# then develop a simple dataframe to compare them
ny_freq_data = pd.DataFrame({'Timeframe': ['Tota... | Change Frequency: 50.28%
| AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
SeverityNow compare severities of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018 & 2017 being in same month-day date range as shut-down period) > | # first find percentages of severity ranks to total accidents for
# total and for each period
ny_sp_ttl = severity_percent(ny_ttl)
ny_sp_sd = severity_percent(ny_sd)
ny_sp_2019 = severity_percent(ny_2019)
ny_sp_2018 = severity_percent(ny_2018)
ny_sp_2017 = severity_percent(ny_2017)
# then again create a simple da... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
NY Takeaways:- Frequency: See an increase in shutdown period, although there was also an even larger jump from 2018 to 2019- Severity: See increase in edge severities (1 & 2) - PennsylvaniaFirst create periods we'll want to compare for the state > | # full state period
pa_ttl = accident_covid[(accident_covid.State == 'PA')]
# shutdown period
pa_sd = accident_covid[
(accident_covid.State == 'PA') &
(accident_covid.Shut_Down == 1)]
# 2019
pa_2019 = accident_covid[(car_accidents_master['Date'] > '2019-03-23') &
(car_accidents_mas... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
FrequencyNow compare frequencies (total number) of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018, & 2017 being in same month-day date range as shut-down period) > | # find total frequencies for state total and in each period
pa_freqs = [pa_ttl.shape[0],
pa_sd.shape[0],
pa_2019.shape[0],
pa_2018.shape[0],
pa_2017.shape[0]
]
# then develop a simple dataframe to compare them
pa_freq_data = pd.DataFrame({'Timeframe': ['Tota... | Change Frequency: 90.76%
| AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
SeverityNow compare severities of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018 & 2017 being in same month-day date range as shut-down period) > | # first find percentages of severity ranks to total accidents for
# total and for each period
pa_sp_ttl = severity_percent(pa_ttl)
pa_sp_sd = severity_percent(pa_sd)
pa_sp_2019 = severity_percent(pa_2019)
pa_sp_2018 = severity_percent(pa_2018)
pa_sp_2017 = severity_percent(pa_2017)
# then again create a simple da... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
PA Takeaways:- Frequency: See a fairly significant increase in accidents during shutdown period- Severity: See increase in severity 1 and drop in severity 2, but not that large - IllinoisFirst create periods we'll want to compare for the state > | # full state period
il_ttl = accident_covid[(accident_covid.State == 'IL')]
# shutdown period
il_sd = accident_covid[
(accident_covid.State == 'IL') &
(accident_covid.Shut_Down == 1)]
# 2019
il_2019 = accident_covid[(car_accidents_master['Date'] > '2019-03-25') &
(car_accidents_mas... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
FrequencyNow compare frequencies (total number) of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018, & 2017 being in same month-day date range as shut-down period) > | # find total frequencies for state total and in each period
il_freqs = [il_ttl.shape[0],
il_sd.shape[0],
il_2019.shape[0],
il_2018.shape[0],
il_2017.shape[0]
]
# then develop a simple dataframe to compare them
il_freq_data = pd.DataFrame({'Timeframe': ['Tota... | Change Frequency: 24.55%
| AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
SeverityNow compare severities of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018 & 2017 being in same month-day date range as shut-down period) > | # first find percentages of severity ranks to total accidents for
# total and for each period
il_sp_ttl = severity_percent(il_ttl)
il_sp_sd = severity_percent(il_sd)
il_sp_2019 = severity_percent(il_2019)
il_sp_2018 = severity_percent(il_2018)
il_sp_2017 = severity_percent(il_2017)
# then again create a simple da... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
IL Takeaways:- Frequency: Increase during shutdown period- Severity: Large increase in severity 3 and decrese in severity 2. Also increase in severity 1 - VirginiaFirst create periods we'll want to compare for the state > | # full state period
va_ttl = accident_covid[(accident_covid.State == 'VA')]
# shutdown period
va_sd = accident_covid[
(accident_covid.State == 'VA') &
(accident_covid.Shut_Down == 1)]
# 2019
va_2019 = accident_covid[(car_accidents_master['Date'] > '2019-03-30') &
(car_accidents_mas... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
FrequencyNow compare frequencies (total number) of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018, & 2017 being in same month-day date range as shut-down period) > | # find total frequencies for state total and in each period
va_freqs = [va_ttl.shape[0],
va_sd.shape[0],
va_2019.shape[0],
va_2018.shape[0],
va_2017.shape[0]
]
# then develop a simple dataframe to compare them
va_freq_data = pd.DataFrame({'Timeframe': ['Tota... | Change Frequency: 113.11%
| AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
SeverityNow compare severities of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018 & 2017 being in same month-day date range as shut-down period) > | # first find percentages of severity ranks to total accidents for
# total and for each period
va_sp_ttl = severity_percent(va_ttl)
va_sp_sd = severity_percent(va_sd)
va_sp_2019 = severity_percent(va_2019)
va_sp_2018 = severity_percent(va_2018)
va_sp_2017 = severity_percent(va_2017)
# then again create a simple da... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
VA Takeaways:- Frequency: See a large increase in frequency during shutdown- Severity: Significant increase in severity 1 and drop in severity 3 - MichiganFirst create periods we'll want to compare for the state > | # full state period
mi_ttl = accident_covid[(accident_covid.State == 'MI')]
# shutdown period
mi_sd = accident_covid[
(accident_covid.State == 'MI') &
(accident_covid.Shut_Down == 1)]
# 2019
mi_2019 = accident_covid[(car_accidents_master['Date'] > '2019-03-24') &
(car_accidents_mas... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
FrequencyNow compare frequencies (total number) of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018, & 2017 being in same month-day date range as shut-down period) > | # find total frequencies for state total and in each period
mi_freqs = [mi_ttl.shape[0],
mi_sd.shape[0],
mi_2019.shape[0],
mi_2018.shape[0],
mi_2017.shape[0]
]
# then develop a simple dataframe to compare them
mi_freq_data = pd.DataFrame({'Timeframe': ['Tota... | Change Frequency: -51.62%
| AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
SeverityNow compare severities of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018 & 2017 being in same month-day date range as shut-down period) > | # first find percentages of severity ranks to total accidents for
# total and for each period
mi_sp_ttl = severity_percent(mi_ttl)
mi_sp_sd = severity_percent(mi_sd)
mi_sp_2019 = severity_percent(mi_2019)
mi_sp_2018 = severity_percent(mi_2018)
mi_sp_2017 = severity_percent(mi_2017)
# then again create a simple da... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
MI Takeaways:- Frequency: Drop in number of accidents- Severity: Increase in edges of severity 1 & 4 -- most dramatic with severity 4 - GeorgiaFirst create periods we'll want to compare for the state > | # full state period
ga_ttl = accident_covid[(accident_covid.State == 'GA')]
# shutdown period
ga_sd = accident_covid[
(accident_covid.State == 'GA') &
(accident_covid.Shut_Down == 1)]
# 2019
ga_2019 = accident_covid[(car_accidents_master['Date'] > '2019-04-03') &
(car_accidents_mas... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
FrequencyNow compare frequencies (total number) of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018, & 2017 being in same month-day date range as shut-down period) > | # find total frequencies for state total and in each period
ga_freqs = [ga_ttl.shape[0],
ga_sd.shape[0],
ga_2019.shape[0],
ga_2018.shape[0],
ga_2017.shape[0]
]
# then develop a simple dataframe to compare them
ga_freq_data = pd.DataFrame({'Timeframe': ['Tota... | Change Frequency: -13.71%
| AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
SeverityNow compare severities of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018 & 2017 being in same month-day date range as shut-down period) > | # first find percentages of severity ranks to total accidents for
# total and for each period
ga_sp_ttl = severity_percent(ga_ttl)
ga_sp_sd = severity_percent(ga_sd)
ga_sp_2019 = severity_percent(ga_2019)
ga_sp_2018 = severity_percent(ga_2018)
ga_sp_2017 = severity_percent(ga_2017)
# then again create a simple da... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
- OregonFirst create periods we'll want to compare for the state > | # full state period
or_ttl = accident_covid[(accident_covid.State == 'OR')]
# shutdown period
or_sd = accident_covid[
(accident_covid.State == 'OR') &
(accident_covid.Shut_Down == 1)]
# 2019
or_2019 = accident_covid[(car_accidents_master['Date'] > '2019-03-23') &
(car_accidents_mas... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
FrequencyNow compare frequencies (total number) of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018, & 2017 being in same month-day date range as shut-down period) >m | # find total frequencies for state total and in each period
or_freqs = [or_ttl.shape[0],
or_sd.shape[0],
or_2019.shape[0],
or_2018.shape[0],
or_2017.shape[0]
]
# then develop a simple dataframe to compare them
or_freq_data = pd.DataFrame({'Timeframe': ['Tota... | Change Frequency: 187.12%
| AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
SeverityNow compare severities of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018 & 2017 being in same month-day date range as shut-down period) > | # first find percentages of severity ranks to total accidents for
# total and for each period
or_sp_ttl = severity_percent(or_ttl)
or_sp_sd = severity_percent(or_sd)
or_sp_2019 = severity_percent(or_2019)
or_sp_2018 = severity_percent(or_2018)
or_sp_2017 = severity_percent(or_2017)
# then again create a simple da... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
- MinnesotaFirst create periods we'll want to compare for the state > | # full state period
mn_ttl = accident_covid[(accident_covid.State == 'MN')]
# shutdown period
mn_sd = accident_covid[
(accident_covid.State == 'MN') &
(accident_covid.Shut_Down == 1)]
# 2019
mn_2019 = accident_covid[(car_accidents_master['Date'] > '2019-03-27') &
(car_accidents_mas... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
FrequencyNow compare frequencies (total number) of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018, & 2017 being in same month-day date range as shut-down period) > | # find total frequencies for state total and in each period
mn_freqs = [mn_ttl.shape[0],
mn_sd.shape[0],
mn_2019.shape[0],
mn_2018.shape[0],
mn_2017.shape[0]
]
# then develop a simple dataframe to compare them
mn_freq_data = pd.DataFrame({'Timeframe': ['Tota... | Change Frequency: 83.18%
| AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
SeverityNow compare severities of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018 & 2017 being in same month-day date range as shut-down period) > | # first find percentages of severity ranks to total accidents for
# total and for each period
mn_sp_ttl = severity_percent(mn_ttl)
mn_sp_sd = severity_percent(mn_sd)
mn_sp_2019 = severity_percent(mn_2019)
mn_sp_2018 = severity_percent(mn_2018)
mn_sp_2017 = severity_percent(mn_2017)
# then again create a simple da... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
- ArizonaFirst create periods we'll want to compare for the state > | # full state period
az_ttl = accident_covid[(accident_covid.State == 'AZ')]
# shutdown period
az_sd = accident_covid[
(accident_covid.State == 'AZ') &
(accident_covid.Shut_Down == 1)]
# 2019
az_2019 = accident_covid[(car_accidents_master['Date'] > '2019-03-31') &
(car_accidents_mas... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
FrequencyNow compare frequencies (total number) of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018, & 2017 being in same month-day date range as shut-down period) > | # find total frequencies for state total and in each period
az_freqs = [az_ttl.shape[0],
az_sd.shape[0],
az_2019.shape[0],
az_2018.shape[0],
az_2017.shape[0]
]
# then develop a simple dataframe to compare them
az_freq_data = pd.DataFrame({'Timeframe': ['Tota... | Change Frequency: 152.00%
| AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
SeverityNow compare severities of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018 & 2017 being in same month-day date range as shut-down period) > | # first find percentages of severity ranks to total accidents for
# total and for each period
az_sp_ttl = severity_percent(az_ttl)
az_sp_sd = severity_percent(az_sd)
az_sp_2019 = severity_percent(az_2019)
az_sp_2018 = severity_percent(az_2018)
az_sp_2017 = severity_percent(az_2017)
# then again create a simple da... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
- TennesseeFirst create periods we'll want to compare for the state > | # full state period
tn_ttl = accident_covid[(accident_covid.State == 'TN')]
# shutdown period
tn_sd = accident_covid[
(accident_covid.State == 'TN') &
(accident_covid.Shut_Down == 1)]
# 2019
tn_2019 = accident_covid[(car_accidents_master['Date'] > '2019-03-31') &
(car_accidents_mas... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
FrequencyNow compare frequencies (total number) of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018, & 2017 being in same month-day date range as shut-down period) > | # find total frequencies for state total and in each period
tn_freqs = [tn_ttl.shape[0],
tn_sd.shape[0],
tn_2019.shape[0],
tn_2018.shape[0],
tn_2017.shape[0]
]
# then develop a simple dataframe to compare them
tn_freq_data = pd.DataFrame({'Timeframe': ['Tota... | Change Frequency: 2.54%
| AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
SeverityNow compare severities of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018 & 2017 being in same month-day date range as shut-down period) > | # first find percentages of severity ranks to total accidents for
# total and for each period
tn_sp_ttl = severity_percent(tn_ttl)
tn_sp_sd = severity_percent(tn_sd)
tn_sp_2019 = severity_percent(tn_2019)
tn_sp_2018 = severity_percent(tn_2018)
tn_sp_2017 = severity_percent(tn_2017)
# then again create a simple da... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
- WashingtonFirst create periods we'll want to compare for the state > | # full state period
wa_ttl = accident_covid[(accident_covid.State == 'WA')]
# shutdown period
wa_sd = accident_covid[
(accident_covid.State == 'WA') &
(accident_covid.Shut_Down == 1)]
# 2019
wa_2019 = accident_covid[(car_accidents_master['Date'] > '2019-03-23') &
(car_accidents_mas... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
FrequencyNow compare frequencies (total number) of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018, & 2017 being in same month-day date range as shut-down period) > | # find total frequencies for state total and in each period
wa_freqs = [wa_ttl.shape[0],
wa_sd.shape[0],
wa_2019.shape[0],
wa_2018.shape[0],
wa_2017.shape[0]
]
# then develop a simple dataframe to compare them
wa_freq_data = pd.DataFrame({'Timeframe': ['Tota... | Change Frequency: -16.54%
| AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
SeverityNow compare severities of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018 & 2017 being in same month-day date range as shut-down period) > | # first find percentages of severity ranks to total accidents for
# total and for each period
wa_sp_ttl = severity_percent(wa_ttl)
wa_sp_sd = severity_percent(wa_sd)
wa_sp_2019 = severity_percent(wa_2019)
wa_sp_2018 = severity_percent(wa_2018)
wa_sp_2017 = severity_percent(wa_2017)
# then again create a simple da... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
- OhioFirst create periods we'll want to compare for the state > | # full state period
oh_ttl = accident_covid[(accident_covid.State == 'OH')]
# shutdown period
oh_sd = accident_covid[
(accident_covid.State == 'OH') &
(accident_covid.Shut_Down == 1)]
# 2019
oh_2019 = accident_covid[(car_accidents_master['Date'] > '2019-03-22') &
(car_accidents_mas... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
FrequencyNow compare frequencies (total number) of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018, & 2017 being in same month-day date range as shut-down period) > | # find total frequencies for state total and in each period
oh_freqs = [oh_ttl.shape[0],
oh_sd.shape[0],
oh_2019.shape[0],
oh_2018.shape[0],
oh_2017.shape[0]
]
# then develop a simple dataframe to compare them
oh_freq_data = pd.DataFrame({'Timeframe': ['Tota... | Change Frequency: 117.88%
| AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
SeverityNow compare severities of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018 & 2017 being in same month-day date range as shut-down period) > | # first find percentages of severity ranks to total accidents for
# total and for each period
oh_sp_ttl = severity_percent(oh_ttl)
oh_sp_sd = severity_percent(oh_sd)
oh_sp_2019 = severity_percent(oh_2019)
oh_sp_2018 = severity_percent(oh_2018)
oh_sp_2017 = severity_percent(oh_2017)
# then again create a simple da... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
- LouisianaFirst create periods we'll want to compare for the state > | # full state period
la_ttl = accident_covid[(accident_covid.State == 'LA')]
# shutdown period
la_sd = accident_covid[
(accident_covid.State == 'LA') &
(accident_covid.Shut_Down == 1)]
# 2019
la_2019 = accident_covid[(car_accidents_master['Date'] > '2019-03-22') &
(car_accidents_mas... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
FrequencyNow compare frequencies (total number) of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018, & 2017 being in same month-day date range as shut-down period) > | # find total frequencies for state total and in each period
la_freqs = [la_ttl.shape[0],
la_sd.shape[0],
la_2019.shape[0],
la_2018.shape[0],
la_2017.shape[0]
]
# then develop a simple dataframe to compare them
la_freq_data = pd.DataFrame({'Timeframe': ['Tota... | Change Frequency: 13.45%
| AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
SeverityNow compare severities of car accidents for total, shut-down period in 2020, 2019, 2018, and 2017 (2019, 2018 & 2017 being in same month-day date range as shut-down period) > | # first find percentages of severity ranks to total accidents for
# total and for each period
la_sp_ttl = severity_percent(la_ttl)
la_sp_sd = severity_percent(la_sd)
la_sp_2019 = severity_percent(la_2019)
la_sp_2018 = severity_percent(la_2018)
la_sp_2017 = severity_percent(la_2017)
# then again create a simple da... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
- OklahomaFirst create periods we'll want to compare for the state > | # full state period
ok_ttl = accident_covid[(accident_covid.State == 'OK')]
# shutdown period
ok_sd = accident_covid[
(accident_covid.State == 'OK') &
(accident_covid.Shut_Down == 1)]
# 2019
ok_2019 = accident_covid[(car_accidents_master['Date'] > '2019-03-24') &
(car_accidents_mas... | _____no_output_____ | AAL | modeling_and_analysis/accident_covid_analysis.ipynb | cecann10/covid-impact-car-accidents |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.