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

• loss plot

use csv data to plot

• line plot
• bar plot
• radar plot
• tree plot
• custom plot

---

🥰Result

plot

data/Pytorch_models_data.csv	data/PaddleYOLO_models_data.csv	data/MMYOLO_model_data.csv

data/llm_eval_data.csv	data/tree.json

---

🔨Usage

requirement

pip install matplotlib
pip install pandas

mmplot install

git clone [email protected]:isLinXu/model-metrics-plot.git
cd model-metrics-plot

pip install -e .

run

python3 main.py

or use your custom data csv

 python3 main.py -c 'csv_path' -n 'figture_name' -p 'plot_type' -t 'title_name' -x 'xlabel_name' -y 'ylabel_name' -f font_size -g False -v 'value_type' -r 'colors' 

line

python3 main.py -c data/model_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/PaddleYOLO_extra_model_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'

bar

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

radar

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

tree