Description
Guidelines
- Share screen shot on your response
- Share the code and the plots
- Put your name and id number
- Clear mark question number
- Upload Word document
- Insert Cover page Questions Attempted
HW13Tensorflow
Q1 Keras and the MNIST Dataset (Optional)
Summarize all the results for the MNIST Dataset - http://yann.lecun.com/exdb/mnist/
Q2: Analyze the MNIST Dataset with Keras
Show screen shots to show installation Explain your results
Hint – use the following links
https://www.kaggle.com/ritupande/self-tutorial-deep-learning-using-keras/data
Install keras on anaconda - https://anaconda.org/conda-forge/keras
Q3. Redo MNIST Dataset with CNN
Hint – use the following links
https://www.kaggle.com/moghazy/guide-to-cnns-with-data-augmentation-keras
Q4 Use Tensorflow play to provide insights on how Tensorflow works -
•Tensor Flow Playground
Q5. Recommendation system with Tensorflow (Optional)
Go through the following tutorial to do the recommendation system with Tensorflow
https://developers.google.com/machine-learning/recommendation/
Q6 Do the XGBoost Exercise on the Titanic dataset
Install XGBoost on anaconda - https://anaconda.org/conda-forge/xgboost
hint on link -
https://www.kaggle.com/ihopethiswillfi/titanic-survival-prediction-in-python-with-xgboost
Q7 Apply XGBoost to Churn Modelling (same dataset as for ANN from previous week’s exercise)
And compare results to the ANN algorithm
Unformatted Attachment Preview
Purchase answer to see full attachment
Explanation & Answer
Hi, here is your assignment. I also attached the plagiarism result ;). Let me know if you have questions or need more help okay :)
1
HW13 Tensorflow
Name
Date
8/4/19
1
2
Q1 Keras and the MNIST Dataset (Optional)
Summaries:
The database is also widely used for training and testing in the field of machine learning.
Goal-Using camera we have to recognize the handwritten digit/numbers. The image of
handwritten digit/numbers are the input and number corresponding to image is the output. In
training we are teaching the system that what is the number corresponding to given image. In
testing model we are getting number corresponding to given image. Accuracy is around 95%98% means the model is predicting correct number to corresponding image with accuracy 95%
to 98%. There is a construction of model by utilizing high-level Keras API which uses
TensorFlow on the backend. I might want to specify that there are a few high- level TensorFlow
APIs, for example, Layers, Keras, and Estimators which causes us to make neural systems with
high-level information....