model-metrics-plot

model-metrics-plot(mmplot)

English | 简体中文

😎 About

This project is developed based on libraries such as Pandas and Matplotlib, and can be used to draw line graphs of multiple index parameters such as algorithm accuracy and speed of multiple deep learning models.

features

use txt or log file to plot

• [x] loss plot

• [x] line plot
• [x] bar plot
• [x] radar plot
• [x] tree plot
• [x] custom plot

🥰Result

plot

| | | | | ------------------------------------------------------------ | ------------------------------------------------------------ | ------------------------------------------------------------ | | data/Pytorchmodelsdata.csv | data/PaddleYOLOmodelsdata.csv | data/MMYOLOmodel_data.csv |

| | | | |-----------------------------------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------| | data/llmevaldata.csv | data/tree.json | | | | | | | | | |

pip install matplotlib
pip install pandas

git clone git@github.com:isLinXu/model-metrics-plot.git
cd model-metrics-plot

pip install -e .

python3 main.py

python3 main.py -c 'csvpath' -n 'figturename' -p 'plottype' -t 'titlename' -x 'xlabelname' -y 'ylabelname' -f fontsize -g False -v 'valuetype' -r 'colors'

python3 main.py -c data/PaddleYOLOextramodel_data.csv -n 'plot.jpg' -p 'line' -t 'MS COCO Object Detection' -x 'PyTorch FP16 RTX3080(ms/img)' -y 'COCO Mask AP val' -f 10 -v 'mAP' -r '#0000FF'

python3 main.py -c data/MMYOLOmodeldata.csv -p bar

python3 main.py -c data/mllmacceval-csv1029.csv -p radar