Update nncf export

src/anomalib/engine/engine.py

@@ -912,22 +912,22 @@ def export(
         CLI Usage:
             1. To export as a torch ``.pt`` file you can run the following command.
                 ```python
-                anomalib export --model Padim --export_mode torch --ckpt_path <PATH_TO_CHECKPOINT>
+                anomalib export --model Padim --export_type torch --ckpt_path <PATH_TO_CHECKPOINT>
                 ```
             2. To export as an ONNX ``.onnx`` file you can run the following command.
                 ```python
-                anomalib export --model Padim --export_mode onnx --ckpt_path <PATH_TO_CHECKPOINT> \
+                anomalib export --model Padim --export_type onnx --ckpt_path <PATH_TO_CHECKPOINT> \
                 --input_size "[256,256]"
                 ```
             3. To export as an OpenVINO ``.xml`` and ``.bin`` file you can run the following command.
                 ```python
-                anomalib export --model Padim --export_mode openvino --ckpt_path <PATH_TO_CHECKPOINT> \
-                --input_size "[256,256] --compression_type "fp16"
+                anomalib export --model Padim --export_type openvino --ckpt_path <PATH_TO_CHECKPOINT> \
+                --input_size "[256,256] --compression_type FP16
                 ```
             4. You can also quantize OpenVINO model with the following.
                 ```python
-                anomalib export --model Padim --export_mode openvino --ckpt_path <PATH_TO_CHECKPOINT> \
-                --input_size "[256,256]" --compression_type "int8_ptq" --data MVTec
+                anomalib export --model Padim --export_type openvino --ckpt_path <PATH_TO_CHECKPOINT> \
+                --input_size "[256,256]" --compression_type INT8_PTQ --data MVTec
                 ```
         """
         export_type = ExportType(export_type)

src/anomalib/models/components/base/export_mixin.py

@@ -330,13 +330,14 @@
         if datamodule is None:
             msg = "Datamodule must be provided for OpenVINO INT8_PTQ compression"
             raise ValueError(msg)
+        datamodule.setup("fit")
 
         model_input = model.input(0)
 
         if model_input.partial_shape[0].is_static:
             datamodule.train_batch_size = model_input.shape[0]
 
-        dataloader = datamodule.train_dataloader()
+        dataloader = datamodule.val_dataloader()
         if len(dataloader.dataset) < 300:
             logger.warning(
                 f">300 images recommended for INT8 quantization, found only {len(dataloader.dataset)} images",
@@ -373,6 +374,8 @@
         if datamodule is None:
             msg = "Datamodule must be provided for OpenVINO INT8_PTQ compression"
             raise ValueError(msg)
+        datamodule.setup("fit")
+
         if metric is None:
             msg = "Metric must be provided for OpenVINO INT8_ACQ compression"
             raise ValueError(msg)
@@ -383,14 +386,14 @@
             datamodule.train_batch_size = model_input.shape[0]
             datamodule.eval_batch_size = model_input.shape[0]
 
-        dataloader = datamodule.train_dataloader()
+        dataloader = datamodule.val_dataloader()
         if len(dataloader.dataset) < 300:
             logger.warning(
                 f">300 images recommended for INT8 quantization, found only {len(dataloader.dataset)} images",
             )
 
         calibration_dataset = nncf.Dataset(dataloader, lambda x: x["image"])
-        validation_dataset = nncf.Dataset(datamodule.val_dataloader())
+        validation_dataset = nncf.Dataset(datamodule.test_dataloader())
 
         if isinstance(metric, str):
             metric = create_metric_collection([metric])[metric]