s444463/ium_444463
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

r

Browse Source
This commit is contained in:
Mikołaj Pokrywka 2022-04-03 18:47:33 +02:00
parent f5528b9b8d
commit 36c6deaeea
2 changed files with 6 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
Dockerfile
View File

@ -10,7 +10,7 @@ RUN pip3 install --upgrade pip
RUN pip3 install pandas RUN pip3 install pandas
RUN pip3 install numpy RUN pip3 install numpy
RUN pip3 install kaggle RUN pip3 install kaggle
RUN apt-get install zip unzip
ARG CUTOFF ARG CUTOFF
ARG KAGGLE_USERNAME ARG KAGGLE_USERNAME
ARG KAGGLE_KEY ARG KAGGLE_KEY
@ -27,4 +27,4 @@ COPY ./process_data.sh .
COPY ./download_data_and_process.py . COPY ./download_data_and_process.py .
COPY ./stats.py . COPY ./stats.py .
# RUN ./process_data.sh RUN ./process_data.sh

8
download_data_and_process.py
View File

@ -1,17 +1,17 @@
import subprocess import subprocess
import pandas as pd import pandas as pd
import numpy as np import numpy as np
import kaggle # import kaggle
kaggle.api.authenticate() # kaggle.api.authenticate()
kaggle.api.dataset_download_files('shivamb/real-or-fake-fake-jobposting-prediction', path='fake_job_postings.csv', unzip=True) # kaggle.api.dataset_download_files('shivamb/real-or-fake-fake-jobposting-prediction', path='fake_job_postings.csv', unzip=True)
data=pd.read_csv('fake_job_postings.csv/fake_job_postings.csv') data=pd.read_csv('fake_job_postings.csv')
data = data.replace(np.nan, '', regex=True) data = data.replace(np.nan, '', regex=True)
print("="*20) print("="*20)