# Example of Research Notes
This is an example of how you might want to organize your research notes.
## Week 1: Jan 26th - 31st, 2022
**Weekly Summary** You can start by summarizing the high level goals of the week. For exmple, "this week, we want to test the sensitivity of our model to changes in hyperparameter as well as identify the weird bug in the plotting function."
Then state your findings from this week, e.g. "we found that our model does not appear to be sensitive to hyperparameter changes, but the performance is overall not good."
**Questions for Group Meeting** Here you want to list questions you have or things you want feedback on:
- How can we explain the poor performance of model? Why isn't there a set of hyperparameters that makes the performance good?
- Can someone look at our plotting function, since we still can't find the bug.
**Feedback from Group Meeting**Here you want to summarize the feedback that you got from the group meeting on the above questions:
- We should look at the model architecture, maybe we are underfitting.
- Send reminder to set up pair programming meeting with so-and-so
### Experiments
Here you describe the experiments you designed to meet the high level goals.
1. Experiments for sensitivity testing:
- **Experiment 1:** give a high level description of what this experiment is suppose to test.
**Set-up:** Describe the experiment set-up
**Evaluation:** Describe how you are summarizing the results and evaluating the experiment
2. Experiments for debugging the plotting funciton:
- **Experiment 1:** give a high level description of what this experiment is suppose to test.
**Set-up:** Describe the experiment set-up
**Evaluation:** Describe how you are summarizing the results and evaluating the experiment
### Results and Analysis
In this section, you summarize the experimental results and provide your analysis. First give an overall high level summary of what the results show. For example, "the results indicate that our models are not sensitive to hyperparamter changes but overall the predictive performance is underwhleming."
#### Experiment 1
**Hypothesis:** State the purpose of this experiment, e.g. what hypothesis are you testing.
Describe in words the results from this experiment. Then high-light which part of the graph or table the reader should look at. Then include the relevant table or graph.

**Analysis:** In what ways do the results support or invalidate the hypothesis. Is there anything unexpected about these results?
#### Overall Conclusion
You've probably done multiple experiments, what do the results imply when considered together? Do they support the same hypothesis or different ones? What are some next questions you might ask given your insights from these experiments?