Update configuration

guides/careamist_api/careamist_api.py

@@ -1,6 +1,6 @@
 #!/usr/bin/env python
 # %%
-# --8<-- [start:careamist_api]
+# --8<-- [start:quick_start_n2v]
 import numpy as np
 from careamics import CAREamist
 from careamics.config import create_n2v_configuration
@@ -25,4 +25,70 @@
 # once trained, predict
 pred_data = np.random.randint(0, 255, (128, 128)).astype(np.float32)
 predction = careamist.predict(source=pred_data)
-# --8<-- [end:careamist_api]
+# --8<-- [end:quick_start_n2v]
+
+# --8<-- [start:quick_start_care]
+import numpy as np
+from careamics import CAREamist
+from careamics.config import create_care_configuration
+
+# create a configuration
+config = create_care_configuration(
+    experiment_name="care_2D",
+    data_type="array",
+    axes="SYX",
+    patch_size=[64, 64],
+    batch_size=1,
+    num_epochs=1,  # (1)!
+)
+
+# instantiate a careamist
+careamist = CAREamist(config)
+
+# train the model
+train_data = np.random.randint(0, 255, (5, 256, 256)).astype(np.float32)  # (2)!
+train_target = np.random.randint(0, 255, (5, 256, 256)).astype(np.float32)
+val_data = np.random.randint(0, 255, (2, 256, 256)).astype(np.float32)
+val_target = np.random.randint(0, 255, (2, 256, 256)).astype(np.float32)
+careamist.train(
+    train_source=train_data,
+    train_target=train_target,
+    val_source=val_data,
+    val_target=val_target,
+)
+
+# once trained, predict
+pred_data = np.random.randint(0, 255, (128, 128)).astype(np.float32)
+predction = careamist.predict(source=pred_data, axes="YX")
+# --8<-- [end:quick_start_care]
+
+# --8<-- [start:quick_start_n2n]
+import numpy as np
+from careamics import CAREamist
+from careamics.config import create_n2n_configuration
+
+# create a configuration
+config = create_n2n_configuration(
+    experiment_name="n2n_2D",
+    data_type="array",
+    axes="YX",
+    patch_size=[64, 64],
+    batch_size=1,
+    num_epochs=1,  # (1)!
+)
+
+# instantiate a careamist
+careamist = CAREamist(config)
+
+# train the model
+train_data = np.random.randint(0, 255, (256, 256)).astype(np.float32)  # (2)!
+train_target = np.random.randint(0, 255, (256, 256)).astype(np.float32)
+careamist.train(
+    train_source=train_data,
+    train_target=train_target,
+)
+
+# once trained, predict
+pred_data = np.random.randint(0, 255, (128, 128)).astype(np.float32)
+predction = careamist.predict(source=pred_data)
+# --8<-- [end:quick_start_care]

guides/careamist_api/configuration/advanced_configuration.py

@@ -11,35 +11,3 @@
     num_epochs=20,
 )
 # --8<-- [end:data]
-
-# %%
-# --8<-- [start:model]
-from careamics.config import FCNAlgorithmConfig, register_model
-from torch import nn, ones
-
-
-@register_model(name="linear_model")  # (1)!
-class LinearModel(nn.Module):
-    def __init__(self, in_features, out_features, *args, **kwargs):
-        super().__init__()
-        self.in_features = in_features
-        self.out_features = out_features
-        self.weight = nn.Parameter(ones(in_features, out_features))
-        self.bias = nn.Parameter(ones(out_features))
-
-    def forward(self, input):
-        return (input @ self.weight) + self.bias
-
-
-config = FCNAlgorithmConfig(
-    algorithm_type="fcn",
-    algorithm="custom",  # (2)!
-    loss="mse",
-    model={
-        "architecture": "custom",  # (3)!
-        "name": "linear_model",  # (4)!
-        "in_features": 10,
-        "out_features": 5,
-    },
-)
-# --8<-- [end:model]

guides/careamist_api/configuration/build_configuration.py

@@ -1,93 +1,138 @@
 #!/usr/bin/env python
-# %%
-# --8<-- [start:as_dict]
-from careamics import Configuration
-
-config_as_dict = {
-    "experiment_name": "my_experiment",  # (1)!
-    "algorithm_config": {  # (2)!
-        "algorithm_type": "fcn",
-        "algorithm": "n2v",
-        "loss": "n2v",
-        "model": {  # (3)!
-            "architecture": "UNet",
-        },
-    },
-    "data_config": {  # (4)!
-        "data_type": "array",
-        "patch_size": [128, 128],
-        "axes": "YX",
-    },
-    "training_config": {
-        "num_epochs": 1,
-    },
-}
-config = Configuration(**config_as_dict)  # (5)!
-# --8<-- [end:as_dict]
 
 # %%
 # --8<-- [start:pydantic]
-from careamics import Configuration
 from careamics.config import (  # (1)!
-    DataConfig,
-    FCNAlgorithmConfig,
+    N2VAlgorithm,
+    N2VConfiguration,
+    N2VDataConfig,
     TrainingConfig,
+    configuration_factory,
 )
 from careamics.config.architectures import UNetModel
+from careamics.config.callback_model import CheckpointModel, EarlyStoppingModel
 from careamics.config.support import (
-    SupportedAlgorithm,
-    SupportedArchitecture,
     SupportedData,
     SupportedLogger,
-    SupportedLoss,
-    SupportedTransform,
 )
-from careamics.config.transformations import N2VManipulateModel
-
-experiment_name = "Pydantic N2V2 example"
-
-# build AlgorithmConfig for the fully convolutional network
-algorithm_model = FCNAlgorithmConfig(  # (2)!
-    algorithm_type="fcn",
-    algorithm=SupportedAlgorithm.N2V.value,  # (3)!
-    loss=SupportedLoss.N2V.value,
-    model=UNetModel(  # (4)!
-        architecture=SupportedArchitecture.UNET.value,
-        in_channels=1,
-        num_classes=1,
-    ),
+from careamics.config.transformations import N2VManipulateModel, XYFlipModel
+
+experiment_name = "N2V_example"
+
+# build the model and algorithm configurations
+model = UNetModel(
+    architecture="UNet",  # (2)!
+    num_channels_init=64,  # (3)!
+    depth=3,
+    # (4)!
+)
+
+algorithm_model = N2VAlgorithm(  # (5)!
+    model=model,
+    # (6)!
 )
 
-# then the DataConfig
-data_model = DataConfig(
+# then the N2VDataConfig
+data_model = N2VDataConfig(  # (7)!
     data_type=SupportedData.ARRAY.value,
     patch_size=(256, 256),
     batch_size=8,
     axes="YX",
-    transforms=[
-        {  # (5)!
-            "name": SupportedTransform.XY_FLIP.value,
-        },
-        N2VManipulateModel(  # (6)!
-            masked_pixel_percentage=0.15,
-        ),
-    ],
-    dataloader_params={  # (7)!
+    transforms=[XYFlipModel(flip_y=False), N2VManipulateModel()],  # (8)!  # (9)!
+    dataloader_params={  # (10)!
         "num_workers": 4,
     },
 )
 
 # then the TrainingConfig
+earlystopping = EarlyStoppingModel(
+    # (11)!
+)
+
+checkpoints = CheckpointModel(every_n_epochs=10)  # (12)!
+
 training_model = TrainingConfig(
     num_epochs=30,
     logger=SupportedLogger.WANDB.value,
+    early_stopping_callback=earlystopping,
+    checkpoint_callback=checkpoints,
+    # (13)!
 )
 
 # finally, build the Configuration
-config = Configuration(  # (8)!
+config = N2VConfiguration(  # (14)!
     experiment_name=experiment_name,
     algorithm_config=algorithm_model,
     data_config=data_model,
     training_config=training_model,
 )
+
+# alternatively, use the factory method
+config2 = configuration_factory(  # (15)!
+    {
+        "experiment_name": experiment_name,
+        "algorithm_config": algorithm_model,
+        "data_config": data_model,
+        "training_config": training_model,
+    }
+)
 # --8<-- [end:pydantic]
+
+if config != config2:
+    raise ValueError("Configurations are not equal (Pydantic).")
+
+# %%
+# --8<-- [start:as_dict]
+from careamics.config import N2VConfiguration, configuration_factory
+
+config_dict = {
+    "experiment_name": "N2V_example",
+    "algorithm_config": {
+        "algorithm": "n2v",  # (1)!
+        "loss": "n2v",
+        "model": {
+            "architecture": "UNet",  # (2)!
+            "num_channels_init": 64,
+            "depth": 3,
+        },
+    },
+    "data_config": {
+        "data_type": "array",
+        "patch_size": [256, 256],
+        "batch_size": 8,
+        "axes": "YX",
+        "transforms": [
+            {
+                "name": "XYFlip",
+                "flip_y": False,
+            },
+            {
+                "name": "N2VManipulate",
+            },
+        ],
+        "dataloader_params": {
+            "num_workers": 4,
+        },
+    },
+    "training_config": {
+        "num_epochs": 30,
+        "logger": "wandb",
+        "early_stopping_callback": {},  # (3)!
+        "checkpoint_callback": {
+            "every_n_epochs": 10,
+        },
+    },
+}
+
+# instantiate specific configuration
+config_as_dict = N2VConfiguration(**config_dict)  # (4)!
+
+# alternatively, use the factory method
+config_as_dict2 = configuration_factory(config_dict)  # (5)!
+# --8<-- [end:as_dict]
+
+if config_as_dict != config_as_dict2:
+    raise ValueError("Configurations are not equal (Dict).")
+
+if config != config_as_dict:
+    raise ValueError("Configurations are not equal (Pydantic vs Dict).")

guides/careamist_api/configuration/configuration_errors.py

@@ -0,0 +1,33 @@
+#!/usr/bin/env python
+# %%
+from careamics.config import create_n2v_configuration
+
+# %%
+# create_n2v_configuration(
+#     experiment_name="N2V_example}",
+#     data_type="array",
+#     axes="YX",
+#     patch_size=[256, 256, 512],
+#     batch_size=8,
+#     num_epochs=30,
+# )
+
+# %%
+create_n2v_configuration(
+    experiment_name="N2V_example",
+    data_type="arrray",
+    axes="YX",
+    patch_size=[256, 256],
+    batch_size=8,
+    num_epochs=30,
+)
+
+# %%
+create_n2v_configuration(
+    experiment_name="N2V_example",
+    data_type="array",
+    axes="YX",
+    patch_size=[256, 256, 512],
+    batch_size=8,
+    num_epochs=30,
+)