Computing Loss in PyTorch

How it works:



How the implementation in PyTorch looks like:
 


Rule of thumb: The more accurate the network, the smaller the loss.


reference: Datacamp's Introduction to Deep Learning with PyTorch

Comments

Post a Comment

Popular posts from this blog

Adding a Dashed Connector in Draw.io

Image
So, over the past two months, I was assigned to create use-case diagrams. I found a handy tool for this purpose i.e.,  draw.io . I never touched this type of diagram in my entire professional career. I learnt how to draw ones when I was in college, sure. But spending most of my time to date as a programmer or data engineer, I never got the chance to actually create one. It was clear that when I got this assignment, I was spending a substantial amount of time just figuring out how to include a dashed connecter  (typically for representing "include" or "extend"). The funny thing is I could not find any clue of how to do this in their blog . So, here's how. Step 0: Initiation Suppose we have two cases: case 1 and case 2 ( I know it was a genuine mistake to name the two cases as use case 1 and use case 2 as shown below. Heck, I'm too lazy to update the pictures, so from now on we will assume the name of the cases are case 1 and case 2, respectively). We want to...
Read more

DBeaver / PostgreSQL: "Error: database already exists" Workaround

Image
Recently, I encountered an error, saying: "Error: database already exists" This error came up when I thought the database I was trying to search was missing. So, I tried to create another new database with the same name. It's quite strange as on the Database Navigator pane at Dbeaver, the database remains hidden somewhere and the annoying this is we cannot create a new one. So, as always, I looked up Stackoverflow and find the workaround was rather simple. Right-click on the connection and fire Edit Connection Head to the PostgreSQL tab and make sure to tick the Show all databases on Click OK Problem solved! Credit to:  oluwasegun  for his response on the following StackOverflow thread:  https://stackoverflow.com/questions/61979762/dbeaver-postgresql-error-database-already-exists-but-i-cant-find-it
Read more

How to Get The Replication Factor of HDFS Files

Image
There are two ways to get the replication factor of HDFS files.  Suppose we want to obtain the replication factor of the 2016 Olympic's Tweet dataset stored in  /data/olympictweets2016rio.  The first way would be to run the following command from your terminal: which will be returning the replication factor as follows: As shown above, the replication factor is 10. However, in most cases, the default replication factor is 3. The other way would be simply using  hadoop fs -ls   command. Just make sure you state the path of the dataset from which you want to obtain the replication factor. The above command will be returning the information of the target directory or files. Have a look at the replication factor represented by the second column right after the permission part. Both ways return the exact replication factor, which in this case is 10. reference:  StackOverflow
Read more