Merge branch 'master' into wyen/fix_pg_update

kubeflow · Apr 26, 2024 · c8d7650 · c8d7650
2 parents 8f4b2e4 + f8f7363
commit c8d7650
Show file tree

Hide file tree

Showing 4 changed files with 17 additions and 20 deletions.
diff --git a/examples/pytorch/language-modeling/train_api_hf_dataset.ipynb b/examples/pytorch/language-modeling/train_api_hf_dataset.ipynb
@@ -21,8 +21,8 @@
     "from kubeflow.storage_initializer.s3 import S3DatasetParams\n",
     "from kubeflow.storage_initializer.hugging_face import (\n",
     "    HuggingFaceModelParams,\n",
-    "    HuggingFaceTrainParams,\n",
-    "    HfDatasetParams,\n",
+    "    HuggingFaceTrainerParams,\n",
+    "    HuggingFaceDatasetParams,\n",
     ")\n",
     "from kubeflow.storage_initializer.constants import INIT_CONTAINER_MOUNT_PATH\n",
     "from peft import LoraConfig\n",
@@ -70,8 +70,8 @@
     "    ),\n",
     "    # it is assumed for text related tasks, you have 'text' column in the dataset.\n",
     "    # for more info on how dataset is loaded check load_and_preprocess_data function in sdk/python/kubeflow/trainer/hf_llm_training.py\n",
-    "    dataset_provider_parameters=HfDatasetParams(repo_id=\"imdatta0/ultrachat_1k\"),\n",
-    "    train_parameters=HuggingFaceTrainParams(\n",
+    "    dataset_provider_parameters=HuggingFaceDatasetParams(repo_id=\"imdatta0/ultrachat_1k\"),\n",
+    "    trainer_parameters=HuggingFaceTrainerParams(\n",
     "        lora_config=LoraConfig(\n",
     "            r=8,\n",
     "            lora_alpha=8,\n",

diff --git a/examples/pytorch/language-modeling/train_api_s3_dataset.ipynb b/examples/pytorch/language-modeling/train_api_s3_dataset.ipynb
@@ -20,9 +20,9 @@
     "from kubeflow.training.api.training_client import TrainingClient\n",
     "from kubeflow.storage_initializer.hugging_face import (\n",
     "    HuggingFaceModelParams,\n",
-    "    HuggingFaceTrainParams,\n",
-    "    HfDatasetParams,\n",
+    "    HuggingFaceTrainerParams,\n",
     ")\n",
+    "from kubeflow.storage_initializer.s3 import S3DatasetParams\n",
     "from kubeflow.storage_initializer.constants import INIT_CONTAINER_MOUNT_PATH\n",
     "from peft import LoraConfig\n",
     "import transformers\n",
@@ -81,16 +81,14 @@
     "    # it is assumed for text related tasks, you have 'text' column in the dataset.\n",
     "    # for more info on how dataset is loaded check load_and_preprocess_data function in sdk/python/kubeflow/trainer/hf_llm_training.py\n",
     "    dataset_provider_parameters=S3DatasetParams(\n",
-    "        {\n",
-    "            \"endpoint_url\": \"http://10.117.63.3\",\n",
-    "            \"bucket_name\": \"test\",\n",
-    "            \"file_key\": \"imdatta0___ultrachat_1k\",\n",
-    "            \"region_name\": \"us-east-1\",\n",
-    "            \"access_key\": s3_access_key,\n",
-    "            \"secret_key\": s3_secret_key,\n",
-    "        }\n",
+    "        endpoint_url=\"http://10.117.63.3\",\n",
+    "        bucket_name=\"test\",\n",
+    "        file_key=\"imdatta0___ultrachat_1k\",\n",
+    "        region_name=\"us-east-1\",\n",
+    "        access_key=s3_access_key,\n",
+    "        secret_key=s3_secret_key,\n",
     "    ),\n",
-    "    train_parameters=HuggingFaceTrainParams(\n",
+    "    trainer_parameters=HuggingFaceTrainerParams(\n",
     "        lora_config=LoraConfig(\n",
     "            r=8,\n",
     "            lora_alpha=8,\n",

diff --git a/examples/pytorch/text-classification/Fine-Tune-BERT-LLM.ipynb b/examples/pytorch/text-classification/Fine-Tune-BERT-LLM.ipynb
@@ -613,8 +613,8 @@
     "from kubeflow.training import TrainingClient\n",
     "from kubeflow.storage_initializer.hugging_face import (\n",
     "    HuggingFaceModelParams,\n",
-    "    HuggingFaceTrainParams,\n",
-    "    HfDatasetParams,\n",
+    "    HuggingFaceTrainerParams,\n",
+    "    HuggingFaceDatasetParams,\n",
     ")\n",
     "\n",
     "import transformers\n",
@@ -646,12 +646,12 @@
     "        \"access_modes\": [\"ReadWriteOnce\"] # Since we use 1 Worker, PVC access mode is ReadWriteOnce.\n",
     "    },\n",
     "    # Use 3000 samples from Yelp dataset.\n",
-    "    dataset_provider_parameters=HfDatasetParams(\n",
+    "    dataset_provider_parameters=HuggingFaceDatasetParams(\n",
     "        repo_id=\"yelp_review_full\",\n",
     "        split=\"train[:3000]\",\n",
     "    ),\n",
     "    # Specify HuggingFace Trainer parameters. In this example, we will skip evaluation and model checkpoints.\n",
-    "    train_parameters=HuggingFaceTrainParams(\n",
+    "    trainer_parameters=HuggingFaceTrainerParams(\n",
     "        training_parameters=transformers.TrainingArguments(\n",
     "            output_dir=\"test_trainer\",\n",
     "            save_strategy=\"no\",\n",

diff --git a/pkg/controller.v1/common/scheduling.go b/pkg/controller.v1/common/scheduling.go
@@ -18,7 +18,6 @@ package common
 
 import (
 	"fmt"
-
 	log "github.com/sirupsen/logrus"
 	k8serrors "k8s.io/apimachinery/pkg/api/errors"
 	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"