These days advancement of technology made possible to solve big calculations in small time, which make currencies like BITcoin, DOGOcoin,etc. possible.
Description
We used a vast training data for training model with tensorflow, keras.and used 80% of data as training data and rest 20% as prediction or verification data. We analysed the steps graphically to monitor what's going beneath the code. We tried to make our model as good or accurate as we could, and in the end the results were as expected.
Getting Started
Dependencies
We have used
Tensorflow
Keras
Seaborn
Scikit-learn
Pandas
Numpy
Matplotlib
Installing
To install above libraries we have to use the following codes in terminal -
Keras - $pip install keras –-user
Tensorflow - $pip install tensorflow -–user
Seaborn - $pip install seaborn –user
Numpy - $pip install numpy --user
Matplotlib - $pip install matplotlib --user
Scikit- Learn - $pip install scikit-learn --user
Pandas - $pip install pandas --user
Executing program
- We installed Jupiter-notebook from https://jupyter.org/
- Use dataset from https://github.com/manishsaini6421/Crypto-Forensic-Bitcoin-/blob/main/Data.csv
- In Jupiter-notebook open codebase file
- Run the code step by step
- Observe the graph after training model
- Wait for expected result.
Help
https://scikit-learn.org/stable/getting_started.html
https://www.tensorflow.org/guide
https://seaborn.pydata.org/tutorial.html
https://keras.io/getting_started/
Authors
Rhythumwinder Singh – [email protected]
Aditya Sharma – [email protected]
Mainsh Kumar Saini – [email protected]
Manmohan – [email protected]
Version History
- 0.1
- Initial Release
License
Apache License 2.0
Acknowledgments
Inspiration, code snippets, etc.
S. Nakamoto, "Bitcoin: A peer-to-peer electronic cash system," 2008.
- T. Dettmers, "Deep learning in a nutshell: Core concepts," NVIDIA Devblogs, 2015.
- D. K. Wind, "Concepts in predictive machine learning," in Maters Thesis, 2014.