Home › Forums › Assignment NPTEL_ › Practical Machine learning with tensor flow › Practical Machine Learning with Tensorflow – Assignment 8
Tagged: Machine Learning, NPTEL, Practical Machine Learning with Tensorflow, Tensorflow, Unit 9  Week 8
 This topic has 0 replies, 1 voice, and was last updated 1 year, 2 months ago by Abhishek Tyagi.

AuthorPosts


April 12, 2020 at 9:25 pm #350Abhishek TyagiKeymaster
1) What will be the values of (z+w) if we execute the following code:
x = tf.constant (0,0)
with tf.GradietTape(persistent=True) as t:
t.watch(x)
Y = tf.cos(x)
Z = tf.sin(x) + tf.tanh(x)
w = z – y + x
dw_dx = t.gradient(w, x)
dz_dx = t.gradient(z, x)
dy_dx = t.gradient(y, x)
Answer 1
2) What will be the values of dz_dx if we execute the previous code:
Answer 2
3) What will be the values of dw_dx if we execute the previous code
Answer 3
4) Please visit this notebook for answering the following questions. What will be the output of the model before training if we
give x = 3.0 as input?Answer 33
5) What is the range of loss after the 10th epoch during the training process?
Answer 1.1 – 1.5
6) What is the range of loss after the 20th epoch?
Answer 1.1 – 1.5
7) What is the range of loss after the 10th epoch if the learning rate was 0.01?
Answer 1525
8) What is the range of loss after the 20th epoch if the learning rate was 0.01?
Answer 515
9) Suppose you have 4 GPUs (having 2GB memory each) to train your deep learning model. You have 1000 data points in your dataset
that takes around 50 GB of storage. What will be the best choice for batch size for one device (using Mirrored Strategy in TF)?Answer 32
10) Select the steps that can be performed in a distributed manner:
Answer Loss computation
Model Prediction.11) Following strategies can be used in case of multiple machines:
Answer MultiWorker Mirrored Strategy
Parameter Servers Strategy This topic was modified 1 year, 2 months ago by Abhishek Tyagi.
 This topic was modified 1 year, 2 months ago by Abhishek Tyagi.
 This topic was modified 1 year, 2 months ago by Abhishek Tyagi.
 This topic was modified 1 year, 2 months ago by Abhishek Tyagi.


AuthorPosts
 You must be logged in to reply to this topic.