Transformer for output data of mat-facebook-downloader for Media Analytics Tool project. The data can be used in mat-facebook-ads-analyzer.
- Python 3
gccandg++libraries
git clone [email protected]:lvtffuk/mat-facebook-ads-transformer.git
cd mat-facebook-ads-transformer
pip install -r requirements.txt
python ./The settings are set with environment variables.
| Variable | Description | Required | Default value |
|---|---|---|---|
INPUT_FILE |
The filepath of the parquet archive from downloader. |
✔️ | |
OUT_DIR |
The directory where the output is stored. | ✔️ | |
CSV_SEPARATOR |
The separator of the input csv files. |
❌ | , |
LANGUAGE |
The language for the udpipe analysis. | ❌ | cs |
The output directory contains models mentioned above and additional files.
| File | Description |
|---|---|
ads_corpus.csv |
The corpus of all text tokens in the input file. |
config.csv |
Configuration. |
df_demographics_unnested.csv |
Ads grouped by demographics data (gender, age). |
df_imp.csv |
Base info about the ad. |
df_region.csv |
Ads grouped by regions. |
total_ads_per_funding.csv |
Total ads per funding. |
total_ads_per_page.csv |
Total ads per page. |
total_region.csv |
Total ads per region. |
The image is stored in GitHub packages registry and the app can be run in the docker environment.
docker pull ghcr.io/lvtffuk/mat-facebook-ads-transformer:latestdocker run \
--name=mat-facebook-ads-transformer \
-e 'INPUT_FILE=./input/archive.parquet' \
-e 'OUTPUT_DIR=./output' \
-v '/absolute/path/to/output/dir:/usr/src/app/output' \
-v '/absolute/path/to/input/dir:/usr/src/app/input' \
ghcr.io/lvtffuk/mat-facebook-ads-transformer:latest The volumes must be set for accessing input and output data.
This work was supported by the European Regional Development Fund-Project “Creativity and Adaptability as Conditions of the Success of Europe in an Interrelated World” (No. CZ.02.1.01/0.0/0.0/16_019/0000734)."
