Build Status Documentation Status Release License



VisualDL is a deep learning visualization tool that can help design deep learning jobs. It includes features such as scalar, parameter distribution, model structure and image visualization. Currently it is being developed at a high pace. New features will be continuously added.

At present, most DNN frameworks use Python as their primary language. VisualDL supports Python by nature. Users can get plentiful visualization results by simply add a few lines of Python code into their model before training.

Besides Python SDK, VisualDL was writen in C++ on the low level. It also provides C++ SDK that can be integrated into other platforms.


VisualDL now provides 4 components:

  • graph
  • scalar
  • image
  • histogram


Graph is compatible with ONNX (Open Neural Network Exchange), Cooperated with Python SDK, VisualDL can be compatible with most major DNN frameworks, including PaddlePaddle, PyTorch and MXNet.


Scalar can be used to show the trends of error during training.


Image can be used to visualize any tensor or intermediate generated image.


Histogram can be used to visualize parameter distribution and trends for any tensor.

Quick Start

To give the VisualDL a quick test, please use the following commands.

# Install the VisualDL. Preferably under a virtual environment.
pip install --upgrade visualdl

# run a demo, vdl_create_scratch_log will create logs for testing.
visualDL --logdir=scratch_log --port=8080

# visit


VisualDL provides both Python SDK and C++ SDK in order to fit more use cases.

Python SDK

VisualDL now supports both Python 2 and Python 3. Below is an example of creating a simple Scalar component and inserting data from different timestamps:

import random
from visualdl import LogWriter

logdir = "./tmp"
logger = LogWriter(logdir, sync_cycle=10000)

# mark the components with 'train' label.
with logger.mode("train"):
    # create a scalar component called 'scalars/scalar0'
    scalar0 = logger.scalar("scalars/scalar0")

# add some records during DL model running.
for step in range(100):
    scalar0.add_record(step, random.random())


Here is the C++ SDK identical to the Python SDK example above:

#include <cstdlib>
#include <string>
#include "visualdl/logic/sdk.h"

namespace vs = visualdl;
namespace cp = visualdl::components;

int main() {
  const std::string dir = "./tmp";
  vs::LogWriter logger(dir, 10000);

  auto tablet = logger.AddTablet("scalars/scalar0");

  cp::Scalar<float> scalar0(tablet);

  for (int step = 0; step < 1000; step++) {
    float v = (float)std::rand() / RAND_MAX;
    scalar0.AddRecord(step, v);

  return 0;

Launch Board

After some logs have been generated during training, users can launch board to see real-time data visualization.

visualDL --logdir <some log dir>

Board also supports the parameters below for remote access:

  • --host set IP
  • --port set port
  • --model_pb specify ONNX format for model file

The VisualDL Graphing system uses GraphViz to visualize the ONNX model. To enable the VisualDL Graph feature, please install GraphViz

How to install from pypi

pip install --upgrade visualdl

How to build and install locally

git clone
cd VisualDL

python bdist_wheel
pip install --upgrade dist/visualdl-*.whl

Run a demo from scratch

# vdl_create_scratch_log is a helper commend that creates mock data.
visualDL --logdir=scratch_log --port=8080

that will start a server locally on port 8080, then you can visit the see the visualdl board.


VisualDL is initially created by PaddlePaddle and ECharts. We welcome everyone to use, comment and contribute to Visual DL :)

VisualDL 相关推荐