# Training

**Example**:

```python
learner = cnn_learner(databunch, pretrained_model, metrics=accuracy)
learn.fit_one_cycle(1,1e-2)
```

## Optimum Learning Rate

Refer to this [link](https://docs.fast.ai/callbacks.lr_finder.html#Suggested-LR)

#### What is Learning Rate?

* Hyper parameter that decides how much gradient should be back propogated
  * i.e. How much we move towards minimum
  * Small learning rate → Converge slowly to minimum
  * Large learning rate → Diverges

![](https://868646840-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-LztfBhQUrZzyA7O_ZkJ%2Fuploads%2Fgit-blob-8700e346f6d3eb1a8d821648bd87841ff0e639c5%2Flearningrate.png?alt=media)

```python
learn.lr_find()
learn.recorder.plot(suggestion=True) # Draws where minimum numerical gradient is (which should be your learning rate)
```

* Choose value that is in the middle of the sharpest downward slope

## The 1cycle Policy

Refer to this [link](https://docs.fast.ai/callbacks.one_cycle.html#What-is-1cycle?)

* Similarly to the regular `learner.fit`, we need to find the optimum learning rate using `lr_finder`.