Create 3d loss surface visualizations, with optimizer path. Issues welcome!

Overview

MLVTK PyPI - Python Version PyPI

A loss surface visualization tool

Png

Simple feed-forward network trained on chess data, using elu activation and Adam optimizer


Gif

Simple feed-forward network trained on chess data, using tanh activation and SGD optimizer


Gif

3 layer feed-forward network trained on hand written letters data, using relu activation, SGD optimizer and learning rate of 2.0. Example of what happens to path when learning rate is too high


Gif

Simple feed-forward network trained on chess data, using hard-sigmoid activation and RMSprop optimizer

Why?

  • :shipit: Simple: A single line addition is all that is needed.
  • Informative: Gain insight into what your model is seeing.
  • 📓 Educational: See how your hyper parameters and architecture impact your models perception.

Quick Start

Requires version
python >= 3.6.1
tensorflow >= 2.3.1
plotly >=4.9.0

Install locally (Also works in google Colab!):

pip install mlvtk

Optionally for use with jupyter notebook/lab:

Notebook

=5.3" "ipywidgets==7.5"">
pip install "notebook>=5.3" "ipywidgets==7.5"

Lab

pip install jupyterlab "ipywidgets==7.5"

# Basic JupyterLab renderer support
jupyter labextension install [email protected]

# OPTIONAL: Jupyter widgets extension for FigureWidget support
jupyter labextension install @jupyter-widgets/jupyterlab-manager [email protected]

Basic Example

from mlvtk.base import Vmodel
import tensorflow as tf
import numpy as np

# NN with 1 hidden layer
inputs = tf.keras.layers.Input(shape=(None,100))
dense_1 = tf.keras.layers.Dense(50, activation='relu')(inputs)
outputs = tf.keras.layers.Dense(10, activation='softmax')(dense_1)
_model = tf.keras.Model(inputs, outputs)

# Wrap with Vmodel
model = Vmodel(_model)
model.compile(optimizer=tf.keras.optimizers.SGD(),
loss=tf.keras.losses.CategoricalCrossentropy(), metrics=['accuracy'])

# All tf.keras.(Model/Sequential/Functional) methods/properties are accessible
# from Vmodel

model.summary()
model.get_config()
model.get_weights()
model.layers

# Create random example data
x = np.random.rand(3, 10, 100)
y = np.random.randint(9, size=(3, 10, 10))
xval = np.random.rand(1, 10, 100)
yval = np.random.randint(9, size=(1,10,10))

# Only difference, model.fit requires validation_data (tf.data.Dataset, or
# other container
history = model.fit(x, y, validation_data=(xval, yval), epochs=10, verbose=0)

# Calling model.surface_plot() returns a plotly.graph_objs.Figure
# model.surface_plot() will attempt to display the figure inline

fig = model.surface_plot()

# fig can save an interactive plot to an html file,
fig.write_html("surface_plot.html")

# or display the plot in jupyter notebook/lab or other compatible tool.
fig.show()
Owner
Research analyst
Jupyter Notebook extension leveraging pandas DataFrames by integrating DataTables and ChartJS.

Jupyter DataTables Jupyter Notebook extension to leverage pandas DataFrames by integrating DataTables JS. About Data scientists and in fact many devel

Marek Čermák 142 Dec 28, 2022
Data aggregated from the reports found at the MCPS COVID Dashboard into a set of visualizations.

Montgomery County Public Schools COVID-19 Visualizer Contents About this project Data Support this project About this project Data All data we use can

James 3 Jan 19, 2022
A way of looking at COVID-19 data that I haven't seen before.

Visualizing Omicron: COVID-19 Deaths vs. Cases Click here for other countries. Data is from Our World in Data/Johns Hopkins University. About this pro

1 Jan 10, 2022
Minimal Ethereum fee data viewer for the terminal, contained in a single python script.

Minimal Ethereum fee data viewer for the terminal, contained in a single python script. Connects to your node and displays some metrics in real-time.

48 Dec 05, 2022
This is a place where I'm playing around with pandas to analyze data in a csv/excel file.

pandas-csv-excel-analysis This is a place where I'm playing around with pandas to analyze data in a csv/excel file. 0-start A very simple cheat sheet

Chuqin 3 Oct 05, 2022
FairLens is an open source Python library for automatically discovering bias and measuring fairness in data

FairLens FairLens is an open source Python library for automatically discovering bias and measuring fairness in data. The package can be used to quick

Synthesized 69 Dec 15, 2022
A napari plugin for visualising and interacting with electron cryotomograms.

napari-tomoslice A napari plugin for visualising and interacting with electron cryotomograms. Installation You can install napari-tomoslice via pip: p

3 Jan 03, 2023
A python script and steps to display locations of peers connected to qbittorrent

A python script (along with instructions) to display the locations of all the peers your qBittorrent client is connected to in a Grafana worldmap dash

62 Dec 07, 2022
PanGraphViewer -- show panenome graph in an easy way

PanGraphViewer -- show panenome graph in an easy way Table of Contents Versions and dependences Desktop-based panGraphViewer Library installation for

16 Dec 17, 2022
Ana's Portfolio

Ana's Portfolio ✌️ Welcome to my Portfolio! You will find here different Projects I have worked on (from scratch) 💪 Projects 💻 1️⃣ Hangman game (Mad

Ana Katherine Cortes Sobrino 9 Mar 15, 2022
Simple addon for snapping active object to mesh ground

Snap to Ground Simple addon for snapping active object to mesh ground How to install: install the Python file as an addon use shortcut "D" in 3D view

Iyad Ahmed 12 Nov 07, 2022
CLAHE Contrast Limited Adaptive Histogram Equalization

A simple code to process images using contrast limited adaptive histogram equalization. Image processing is becoming a major part of data processig.

Happy N. Monday 4 May 18, 2022
Show Data: Show your dataset in web browser!

Show Data is to generate html tables for large scale image dataset, especially for the dataset in remote server. It provides some useful commond line tools and fully customizeble API reference to gen

Dechao Meng 83 Nov 26, 2022
🎨 Python3 binding for `@AntV/G2Plot` Plotting Library .

PyG2Plot 🎨 Python3 binding for @AntV/G2Plot which an interactive and responsive charting library. Based on the grammar of graphics, you can easily ma

hustcc 990 Jan 05, 2023
哔咔漫画window客户端,界面使用PySide2,已实现分类、搜索、收藏夹、下载、在线观看、waifu2x等功能。

picacomic-windows 哔咔漫画window客户端,界面使用PySide2,已实现分类、搜索、收藏夹、下载、在线观看等功能。 功能介绍 登陆分流,还原安卓端的三个分流入口 分类,搜索,排行,收藏夹使用同一的逻辑,滚轮下滑自动加载下一页,双击打开 漫画详情,章节列表和评论列表 下载功能,目

1.8k Dec 31, 2022
OpenStats is a library built on top of streamlit that extracts data from the Github API and shows the main KPIs

Open Stats Discover and share the KPIs of your OpenSource project. OpenStats is a library built on top of streamlit that extracts data from the Github

Pere Miquel Brull 4 Apr 03, 2022
2021 grafana arbitrary file read

2021_grafana_arbitrary_file_read base on pocsuite3 try 40 default plugins of grafana alertlist annolist barchart cloudwatch dashlist elasticsearch gra

ATpiu 5 Nov 09, 2022
Create Badges with stats of Scratch User, Project and Studio. Use those badges in Github readmes, etc.

Scratch-Stats-Badge Create customized Badges with stats of Scratch User, Studio or Project. Use those badges in Github readmes, etc. Examples Document

Siddhesh Chavan 5 Aug 28, 2022
With Holoviews, your data visualizes itself.

HoloViews Stop plotting your data - annotate your data and let it visualize itself. HoloViews is an open-source Python library designed to make data a

HoloViz 2.3k Jan 04, 2023
Create matplotlib visualizations from the command-line

MatplotCLI Create matplotlib visualizations from the command-line MatplotCLI is a simple utility to quickly create plots from the command-line, levera

Daniel Moura 46 Dec 16, 2022