You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
66 lines
2.5 KiB
66 lines
2.5 KiB
2 years ago
|
Both the Ultralytics YOLO command-line and python interfaces are simply a high-level abstraction on the base engine executors. Let's take a look at the Trainer engine.
|
||
|
|
|
||
|
|
## BaseTrainer
|
||
|
|
BaseTrainer contains the generic boilerplate training routine. It can be customized for any task based over overidding the required functions or operations as long the as correct formats are followed. For example you can support your own custom model and dataloder by just overriding these functions:
|
||
|
|
|
||
|
|
* `get_model(cfg, weights)` - The function that builds a the model to be trained
|
||
|
|
* `get_dataloder()` - The function that builds the dataloder
|
||
|
|
More details and source code can be found in [`BaseTrainer` Reference](../reference/base_trainer.md)
|
||
|
|
|
||
|
|
## DetectionTrainer
|
||
|
|
Here's how you can use the YOLOv8 `DetectionTrainer` and customize it.
|
||
|
|
```python
|
||
|
|
from Ultrlaytics.yolo.v8 import DetectionTrainer
|
||
|
|
|
||
|
|
trainer = DetectionTrainer(overrides={...})
|
||
|
|
trainer.train()
|
||
|
|
trained_model = trainer.best # get best model
|
||
|
|
```
|
||
|
|
|
||
|
|
### Customizing the DetectionTrainer
|
||
|
|
Let's customize the trainer **to train a custom detection model** that is not supported directly. You can do this by simply overloading the existing the `get_model` functionality:
|
||
|
|
```python
|
||
|
|
from Ultrlaytics.yolo.v8 import DetectionTrainer
|
||
|
|
|
||
|
|
class CustomTrainer(DetectionTrainer):
|
||
|
|
def get_model(self, cfg, weights):
|
||
|
|
...
|
||
|
|
|
||
|
|
trainer = CustomTrainer(overrides={...})
|
||
|
|
trainer.train()
|
||
|
|
```
|
||
|
|
You now realize that you need to customize the trainer further to:
|
||
|
|
|
||
|
|
* Customize the `loss function`.
|
||
|
|
* Add `callback` that uploads model to your google drive after every 10 `epochs`
|
||
|
|
Here's how you can do it:
|
||
|
|
|
||
|
|
```python
|
||
|
|
from Ultrlaytics.yolo.v8 import DetectionTrainer
|
||
|
|
|
||
|
|
class CustomTrainer(DetectionTrainer):
|
||
|
|
def get_model(self, cfg, weights):
|
||
|
|
...
|
||
|
|
|
||
|
|
def criterion(self, preds, batch):
|
||
|
|
# get ground truth
|
||
|
|
imgs = batch["imgs"]
|
||
|
|
bboxes = batch["bboxes"]
|
||
|
|
...
|
||
|
|
return loss, loss_items # see Reference-> Trainer for details on the expected format
|
||
|
|
|
||
|
|
# callback to upload model weights
|
||
|
|
def log_model(trainer):
|
||
|
|
last_weight_path = trainer.last
|
||
|
|
...
|
||
|
|
|
||
|
|
trainer = CustomTrainer(overrides={...})
|
||
|
|
trainer.add_callback("on_train_epoch_end", log_model) # Adds to existing callback
|
||
|
|
trainer.train()
|
||
|
|
```
|
||
|
|
|
||
|
|
To know more about Callback triggering events and entry point, checkout our Callbacks guide # TODO
|
||
|
|
|
||
|
|
## Other engine components
|
||
|
|
There are other componenets that can be customized similarly like `Validators` and `Predictiors`
|
||
|
|
To know more about their implementation details, go to Reference
|