# Getting back to where you left off!

**URL:** <https://community.wandb.ai/t/getting-back-to-where-you-left-off/754>\
**Category:** Show the Community!\
**Created:** [September 23, 2021, 2:14pm UTC](https://community.wandb.ai/t/getting-back-to-where-you-left-off/754 "2021-09-23T14:14:20Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![aditya-digala-gmail](https://sea2.discourse-cdn.com/flex020/user_avatar/community.wandb.ai/aditya-digala-gmail/32/223_2.png) [@aditya-digala-gmail](https://community.wandb.ai/u/aditya-digala-gmail)\
**Post date:** [September 23, 2021, 2:14pm UTC](https://community.wandb.ai/t/getting-back-to-where-you-left-off/754/1 "2021-09-23T14:14:20Z")

</div>

## I want to ask the community how you deal with project gaps.

I’ve been working on a research project for 3-4 months until I had to pause it to prepare for my semester exams.  
It’s been more than a month or so since I worked on my project. I appreciate suggestions from the community to help me get back to where I left off **asap.!**

---

<div class="post-metadata">

**Author:** ![sahiljuneja](https://sea2.discourse-cdn.com/flex020/user_avatar/community.wandb.ai/sahiljuneja/32/294_2.png) [@sahiljuneja](https://community.wandb.ai/u/sahiljuneja)\
**Post date:** [September 23, 2021, 8:02pm UTC](https://community.wandb.ai/t/getting-back-to-where-you-left-off/754/2 "2021-09-23T20:02:41Z")

</div>

( **Note:** I am making assumptions about your project given the context of the community.)

I haven’t worked on projects that last that long. But, when I study or learn something there are often gaps varying from weeks to even months.

The notes I take during that process is what helps me jump back with relative ease. I also sort of mark off what I did last or where I left off last.

If you haven’t taken any notes or alternatives like

- frequently committing your work to a GitHub repo with good comments, or
- maintaining some list of past experiments and results

then the only thing I can recommend is to just go into experiment mode.

- run your code a bunch of times and play around with whatever you can (like hyperparameters as a simple example) to try and see what you can recall
- glance over some of the research papers you read last to try to jumpstart some of the ideas you might have been considering previously
- something like this helped me once somehow → if feasible, check out your browser history from around that time and try to connect the dots

Maybe some others have better suggestions for this, but I don’t think there’s much to be done if you haven’t maintained much of a record of what you were doing.

It’s not always apt, but does hold at times as per me!

[![](https://us1.discourse-cdn.com/flex020/uploads/wandb/original/1X/e09a8e366c31208ee1ab22276b00d3889d0c0302.jpeg "Adam Savage: Science vs. screwing around.") ](https://www.youtube.com/watch?v=BSUMBBFjxrY)

---

<div class="post-metadata">

**Author:** ![aditya-digala-gmail](https://sea2.discourse-cdn.com/flex020/user_avatar/community.wandb.ai/aditya-digala-gmail/32/223_2.png) [@aditya-digala-gmail](https://community.wandb.ai/u/aditya-digala-gmail)\
**Post date:** [September 24, 2021, 4:32am UTC](https://community.wandb.ai/t/getting-back-to-where-you-left-off/754/3 "2021-09-24T04:32:32Z")

</div>

Thanks for your suggestions.  
As I said, it’s a research-based project, so it is time-consuming.  
I do have notes. And I’ll probably start from there as you suggested!

🙌