add tags copyright and github workflow

.github/workflows/ai-ml-gke-finetuning-gemma.yml

@@ -0,0 +1,21 @@
+name: ai-ml-finetune-gemma
+on:
+  push:
+    branches:
+      - main
+    paths:
+      - '.github/workflows/ai-ml-gke-finetuning-gemma.yml'
+      - 'ai-ml/llm-finetuning-gemma/**'
+  pull_request:
+    paths:
+      - '.github/workflows/ai-ml-gke-finetuning-gemma.yml'
+      - 'ai-ml/llm-finetuning-gemma/**'
+jobs:
+  gke-a100-jax:
+    runs-on: ubuntu-22.04
+    steps:
+      - uses: actions/checkout@v4
+      - name: build container for llm-finetuning-gemma tutorial
+        run: |
+          cd ai-ml/llm-fineturning-gemma/transformers
+          docker build --tag finetune .

ai-ml/llm-finetuning-gemma/Dockerfile

@@ -0,0 +1,29 @@
+# Copyright 2024 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+# [START gke_aiml_llm_finetune_gemma_single_node_docker]
+FROM nvidia/cuda:12.2.0-runtime-ubuntu22.04
+
+RUN apt-get update && \
+    apt-get -y --no-install-recommends install python3-dev gcc python3-pip git && \
+    rm -rf /var/lib/apt/lists/*
+
+RUN pip3 install --no-cache-dir accelerate bitsandbytes datasets transformers peft trl torch
+
+COPY finetune.py /finetune.py
+
+ENV PYTHONUNBUFFERED 1
+
+CMD python3 /finetune.py --device cuda
+# [END gke_aiml_llm_finetune_gemma_single_node_docker]

ai-ml/llm-finetuning-gemma/transformers/finetune.py → ai-ml/llm-finetuning-gemma/finetune.py

@@ -1,3 +1,18 @@
+# Copyright 2024 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+# [START gke_aiml_llm_finetune_gemma_single_node_py]
 import os
 import torch
 from datasets import load_dataset, Dataset
@@ -232,4 +247,5 @@ def transform(data):
 
 model.push_to_hub(new_model, check_pr=True)
 
-tokenizer.push_to_hub(new_model, check_pr=True)
+tokenizer.push_to_hub(new_model, check_pr=True)
+# [END gke_aiml_llm_finetune_gemma_single_node_py]

ai-ml/llm-finetuning-gemma/transformers/finetune.yaml → ai-ml/llm-finetuning-gemma/finetune.yaml

@@ -12,6 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+# [START gke_aiml_llm_finetune_gemma_single_node_yaml]
 apiVersion: batch/v1
 kind: Job
 metadata:
@@ -35,7 +36,7 @@ spec:
         - name: MODEL_NAME
           value: "google/gemma-2b"
         - name: NEW_MODEL
-          value: "gemma-2b-sql-kubecon-eu-2024"
+          value: "gemma-2b-sql-finetuned"
         - name: LORA_R
           value: "8"
         - name: LORA_ALPHA
@@ -66,4 +67,5 @@ spec:
           medium: Memory
       nodeSelector:
         cloud.google.com/gke-accelerator: nvidia-l4
-      restartPolicy: OnFailure
+      restartPolicy: OnFailure
+# [END gke_aiml_llm_finetune_gemma_single_node_yaml]