https://github.com/shambac/shamboflow
Fierce tensorflow competitor
Science Score: 13.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
○CITATION.cff file
-
✓codemeta.json file
Found codemeta.json file -
○.zenodo.json file
-
○DOI references
-
○Academic publication links
-
○Academic email domains
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (15.8%) to scientific vocabulary
Keywords
Repository
Fierce tensorflow competitor
Basic Info
- Host: GitHub
- Owner: ShambaC
- License: mit
- Language: Python
- Default Branch: main
- Homepage: https://pypi.org/project/shamboflow/
- Size: 373 KB
Statistics
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
- Releases: 4
Topics
Metadata Files
README.md
[!IMPORTANT] This is an elaborate meme I put a lot of effort into. So please, no one sue me.
ShamboFlow is an open source API for creating machine learning models. It is only available in python.
ShamboFlow is super fast drop in replacemnet for TensorFlow (Read adds nothing, not even performance improvement). It is build from scratch (Read, using numpy) and comes with Cuda GPU support out of the box. I will tell the story at the end of this file on how this came to be.
On a serious note, I always wanted to implement a neural network using just numpy and no additional libraries and this gave me an excuse to do so. And so I did. I made this in a week. Learned a lot of stuff in the process and it was a stressfull and fun experience. This library is dependent on numpy as stated but also uses cupy to add GPU support. Other two dependencies are tqdm for progress bar and colorama for colorful texts. I will probably work more on this as I have already put in quite some effort.
Documentation
Install
Install using the pip package.
bash
$ pip install shamboflow
To update ShamboFlow to the latest version, add --upgrade flag to the above command
Example
A small example program that shows how to create a simple ANN with 3-2-1 topology and train it with data to perform predictions.
Define the model and train it
```python import numpy as np
Dataset
xdata = np.array([[1, 0, 1]]) ydata = np.array([[1]])
Parameters
learningrate = 0.9 trainepochs = 20
Import the library
import shamboflow as sf
Create a model
model = sf.models.Sequential()
Add layers
model.add(sf.layers.Dense(3)) model.add(sf.layers.Dense(2, activation='sigmoid')) model.add(sf.layers.Dense(1, activation='sigmoid'))
Compile the model
model.compile(learningrate=learningrate, loss='meansquarederror', verbose=True)
Callbacks
checkpoint = sf.callbacks.ModelCheckpoint(monitor='loss', savebestonly=True, verbose=True)
Train the model with the dataset
model.fit( xdata, ydata, epochs=train_epochs, callbacks=[checkpoint] )
Save the trained model to disk
model.save('model.meow') ```
Load the saved model and predict
```python import numpy as np import shamboflow as sf
model = sf.models.load_model("./model.meow")
a = np.array([[1, 0, 1]])
res = model.predict(a) print(res) ```
Story
Its storytime.
Last week we had a class on Neural Networks at university. At the end of the class, our professor told us to implement the given network in python. Now, previously he had told us to not use any libraries to perform our tasks as that would just ruin the purpose of learning algorithms. So, I got excited that I am gonna implement a neural network using just python. Then he told us that we can use libraries for making the network. And I was a little bummed. My friend jokingly told me that, "No you have to make it". And I said, if I finish it within a week, will you use it in the assignment. My friends agreed to it.
So, here it is. My library. I am so gonna make them use this for the assignments.
Owner
- Name: Shamba Chowdhury
- Login: ShambaC
- Kind: user
- Location: Penthouse in Sundarban
- Repositories: 1
- Profile: https://github.com/ShambaC
Nyanpasu ! I like to read manga and do random stuffs. 😛
GitHub Events
Total
- Release event: 2
- Push event: 3
- Create event: 2
Last Year
- Release event: 2
- Push event: 3
- Create event: 2
Issues and Pull Requests
Last synced: 10 months ago
All Time
- Total issues: 0
- Total pull requests: 0
- Average time to close issues: N/A
- Average time to close pull requests: N/A
- Total issue authors: 0
- Total pull request authors: 0
- Average comments per issue: 0
- Average comments per pull request: 0
- Merged pull requests: 0
- Bot issues: 0
- Bot pull requests: 0
Past Year
- Issues: 0
- Pull requests: 0
- Average time to close issues: N/A
- Average time to close pull requests: N/A
- Issue authors: 0
- Pull request authors: 0
- Average comments per issue: 0
- Average comments per pull request: 0
- Merged pull requests: 0
- Bot issues: 0
- Bot pull requests: 0
Top Authors
Issue Authors
Pull Request Authors
Top Labels
Issue Labels
Pull Request Labels
Packages
- Total packages: 1
-
Total downloads:
- pypi 25 last-month
- Total dependent packages: 0
- Total dependent repositories: 0
- Total versions: 5
- Total maintainers: 1
pypi.org: shamboflow
A fierce Tensorflow competitor
- Homepage: https://github.com/ShambaC/shamboflow
- Documentation: https://shamboflow.readthedocs.io/
- License: MIT License
-
Latest release: 1.0.2
published over 1 year ago