renaming folder to storage initalizer

.github/workflows/publish-core-images.yaml

@@ -24,3 +24,6 @@ jobs:
             dockerfile: build/images/training-operator/Dockerfile
           - component-name: kubectl-delivery
             dockerfile: build/images/kubectl-delivery/Dockerfile
+          - component-name: storage-initializer
+             dockerfile: sdk/python/kubeflow/storage_initializer/Dockerfile
+             context: sdk/python/kubeflow/storage_initializer

sdk/python/kubeflow/storage_init_container/Dockerfile → sdk/python/kubeflow/storage_initializer/Dockerfile

@@ -5,7 +5,7 @@ FROM python:3.11
 WORKDIR /app
 
 # Copy the Python package and its source code into the container
-COPY . /app/storage_init_container
+COPY . /app/storage_initializer
 
 # Copy the requirements.txt file into the container
 COPY requirements.txt /app/requirements.txt
@@ -14,4 +14,4 @@ COPY requirements.txt /app/requirements.txt
 RUN pip install --no-cache-dir -r requirements.txt
 
 # Run storage.py when the container launches
-ENTRYPOINT ["python", "-m", "storage_init_container.storage"]
+ENTRYPOINT ["python", "-m", "storage_initializer.storage"]

sdk/python/kubeflow/storage_initializer/constants.py

@@ -0,0 +1,3 @@
+INIT_CONTAINER_MOUNT_PATH = "/workspace"
+VOLUME_PATH_DATASET = INIT_CONTAINER_MOUNT_PATH + "/dataset"
+VOLUME_PATH_MODEL = INIT_CONTAINER_MOUNT_PATH + "/model"

sdk/python/kubeflow/storage_init_container/hugging_face.py → sdk/python/kubeflow/storage_initializer/hugging_face.py

@@ -4,7 +4,7 @@
 from urllib.parse import urlparse
 import json, os
 from typing import Union
-from .constants import INIT_CONTAINER_MOUNT_PATH
+from .constants import VOLUME_PATH_DATASET, VOLUME_PATH_MODEL
 from .abstract_model_provider import modelProvider
 from .abstract_dataset_provider import datasetProvider
 
@@ -24,21 +24,12 @@ class HuggingFaceModelParams:
     model_uri: str
     transformer_type: TRANSFORMER_TYPES
     access_token: str = None
-    download_dir: str = field(default=os.path.join(INIT_CONTAINER_MOUNT_PATH, "models"))
 
     def __post_init__(self):
         # Custom checks or validations can be added here
         if self.model_uri == "" or self.model_uri is None:
             raise ValueError("model_uri cannot be empty.")
 
-    @property
-    def download_dir(self):
-        return self.download_dir
-
-    @download_dir.setter
-    def download_dir(self, value):
-        raise AttributeError("Cannot modify read-only field 'download_dir'")
-
 
 @dataclass
 class HuggingFaceTrainParams:
@@ -62,35 +53,24 @@ def download_model_and_tokenizer(self):
         transformer_type_class.from_pretrained(
             self.model,
             token=self.config.access_token,
-            cache_dir=self.config.download_dir,
+            cache_dir=VOLUME_PATH_MODEL,
             trust_remote_code=True,
         )
         transformers.AutoTokenizer.from_pretrained(
-            self.model, cache_dir=self.config.download_dir
+            self.model, cache_dir=VOLUME_PATH_MODEL
         )
 
 
 @dataclass
 class HfDatasetParams:
     repo_id: str
     access_token: str = None
-    download_dir: str = field(
-        default=os.path.join(INIT_CONTAINER_MOUNT_PATH, "datasets")
-    )
 
     def __post_init__(self):
         # Custom checks or validations can be added here
         if self.repo_id == "" or self.repo_id is None:
             raise ValueError("repo_id is None")
 
-    @property
-    def download_dir(self):
-        return self.download_dir
-
-    @download_dir.setter
-    def download_dir(self, value):
-        raise AttributeError("Cannot modify read-only field 'download_dir'")
-
 
 class HuggingFaceDataset(datasetProvider):
     def load_config(self, serialised_args):
@@ -104,4 +84,4 @@ def download_dataset(self):
         if self.config.access_token:
             huggingface_hub.login(self.config.access_token)
 
-        load_dataset(self.config.repo_id, cache_dir=self.config.download_dir)
+        load_dataset(self.config.repo_id, cache_dir=VOLUME_PATH_DATASET)

sdk/python/kubeflow/storage_init_container/s3.py → sdk/python/kubeflow/storage_initializer/s3.py

@@ -3,7 +3,7 @@
 import boto3
 from urllib.parse import urlparse
 from .abstract_dataset_provider import datasetProvider
-from .constants import INIT_CONTAINER_MOUNT_PATH
+from .constants import VOLUME_PATH_DATASET
 
 
 @dataclass
@@ -14,9 +14,6 @@ class S3DatasetParams:
     region_name: str = None
     access_key: str = None
     secret_key: str = None
-    download_dir: str = field(
-        default=os.path.join(INIT_CONTAINER_MOUNT_PATH, "datasets")
-    )
 
     def is_valid_url(self, url):
         try:
@@ -36,14 +33,6 @@ def __post_init__(self):
             raise ValueError("bucket_name or endpoint_url or file_key is None")
         self.is_valid_url(self.endpoint_url)
 
-    @property
-    def download_dir(self):
-        return self.download_dir
-
-    @download_dir.setter
-    def download_dir(self, value):
-        raise AttributeError("Cannot modify read-only field 'download_dir'")
-
 
 class S3(datasetProvider):
     def load_config(self, serialised_args):
@@ -63,6 +52,6 @@ def download_dataset(self):
         s3_client.download_file(
             self.config.bucket_name,
             self.config.file_key,
-            os.path.join(self.config.download_dir, self.config.file_key),
+            os.path.join(VOLUME_PATH_DATASET, self.config.file_key),
         )
-        print(f"File downloaded to: {self.config.download_dir}")
+        print(f"File downloaded to: {VOLUME_PATH_DATASET}")

sdk/python/kubeflow/training/api/training_client.py

@@ -24,7 +24,11 @@
 from kubeflow.training.api_client import ApiClient
 from kubeflow.training.constants import constants
 from kubeflow.training.utils import utils
-from kubeflow.storage_init_container.constants import INIT_CONTAINER_MOUNT_PATH
+from kubeflow.storage_initializer.constants import (
+    INIT_CONTAINER_MOUNT_PATH,
+    VOLUME_PATH_DATASET,
+    VOLUME_PATH_MODEL,
+)
 
 
 logger = logging.getLogger(__name__)
@@ -116,8 +120,8 @@ def train(
             print(
                 "train api dependencies not installed. Run pip install -U 'kubeflow-training[huggingface]' "
             )
-        from kubeflow.storage_init_container.s3 import S3DatasetParams
-        from kubeflow.storage_init_container.hugging_face import (
+        from kubeflow.storage_initializer.s3 import S3DatasetParams
+        from kubeflow.storage_initializer.hugging_face import (
             HuggingFaceModelParams,
             HuggingFaceTrainParams,
             HfDatasetParams,
@@ -209,9 +213,9 @@ def train(
                 "--transformer_type",
                 model_provider_parameters.transformer_type.__class__.__name__,
                 "--model_dir",
-                model_provider_parameters.download_dir,
+                VOLUME_PATH_MODEL,
                 "--dataset_dir",
-                dataset_provider_parameters.download_dir,
+                VOLUME_PATH_DATASET,
                 "--dataset_name",
                 dataset_provider_parameters.repo_id,
                 "--lora_config",
@@ -222,7 +226,7 @@ def train(
             volume_mounts=[
                 models.V1VolumeMount(
                     name=constants.TRAINER_PV,
-                    mount_path=constants.TRAINER_CONTAINER_MOUNT_PATH,
+                    mount_path=INIT_CONTAINER_MOUNT_PATH,
                 )
             ],
             resources=resources_per_worker,

sdk/python/kubeflow/training/constants/constants.py

@@ -168,5 +168,3 @@
     models.KubeflowOrgV1MPIJob,
     models.KubeflowOrgV1PaddleJob,
 ]
-
-TRAINER_CONTAINER_MOUNT_PATH = "/workspace"