kubeflow

Author	SHA1	Message	Date
Billy D.	d4eb54d92b	pipelines go to gravenhollow now. All checks were successful Compile and Upload Pipelines / Compile & Upload (push) Successful in 15s Details Compile and Upload Pipelines / Notify (push) Successful in 1s Details	2026-02-18 07:14:12 -05:00
Billy D.	7f2b011c95	its coqui-tts not coqui. All checks were successful Compile and Upload Pipelines / Compile & Upload (push) Successful in 17s Details Compile and Upload Pipelines / Notify (push) Successful in 1s Details	2026-02-18 07:11:56 -05:00
Billy D.	9c51355baa	feat: add distributed CPU training pipeline via Ray Train All checks were successful Compile and Upload Pipelines / Compile & Upload (push) Successful in 17s Details Compile and Upload Pipelines / Notify (push) Successful in 2s Details	2026-02-14 11:12:51 -05:00
Billy D.	1e5aa0a6d8	chore: add Renovate config for automated dependency updates Ref: ADR-0057	2026-02-13 15:34:52 -05:00
Billy D.	5c886bf6a5	feat: add voice cloning pipeline (S3 audio → Whisper → VITS training → Gitea) All checks were successful Compile and Upload Pipelines / Compile & Upload (push) Successful in 15s Details Compile and Upload Pipelines / Notify (push) Successful in 1s Details	2026-02-13 10:54:04 -05:00
Billy D.	5cef268efc	ci: re-trigger pipeline upload after networkpolicy fix	2026-02-13 10:40:42 -05:00
Billy D.	fc036b0e72	fix: use find+while loop instead of multiline output in for loop	2026-02-13 10:32:35 -05:00
Billy D.	321eca5943	feat: add QLoRA PDF pipeline and Gitea CI workflow Some checks failed Compile and Upload Pipelines / Compile & Upload (push) Failing after 58s Details Compile and Upload Pipelines / Notify (push) Successful in 1s Details - qlora_pdf_pipeline.py: 6-step QLoRA fine-tuning pipeline (S3 PDFs → prepare data → train → evaluate → push to Gitea → MLflow) - .gitea/workflows/compile-upload.yaml: auto-compile and upload all pipelines to Kubeflow on push, with ntfy notifications	2026-02-13 10:28:53 -05:00
Billy D.	45996a8dbf	feat: add DVD/video transcription pipeline 5-step KFP pipeline: 1. extract_audio: ffmpeg extracts 16kHz mono WAV from DVD/video 2. chunk_audio: splits into 5-minute segments for Whisper 3. transcribe_chunks: sends each chunk to Whisper STT endpoint 4. format_transcript: produces SRT, VTT, or TXT with timestamps 5. log_metrics: logs run to MLflow (dvd-transcription experiment)	2026-02-13 09:22:56 -05:00
Billy D.	bc4b230dd9	feat: add vLLM tuning pipeline + recompile voice pipelines with MLflow New: - vllm_tuning_pipeline.py: A/B benchmark different vLLM configs, logs latency/TPS/TTFT to MLflow (vllm-tuning experiment) - vllm_tuning_pipeline.yaml: compiled KFP YAML Updated: - voice_pipeline.py: per-step NamedTuple outputs with latency tracking, new log_pipeline_metrics MLflow component - voice_pipeline.yaml, tts_pipeline.yaml, rag_pipeline.yaml: recompiled	2026-02-13 08:24:11 -05:00
Billy D.	cee21f124c	feat: add MLflow tracking to evaluation pipeline - Add create_mlflow_run and log_evaluation_to_mlflow KFP components - Log accuracy, correct/total counts, pass/fail to MLflow experiment - Upload evaluation_results.json as artifact - Wire MLflow run into pipeline DAG before NATS publish	2026-02-12 06:15:13 -05:00
Billy D.	bd8c8616d0	updates.	2026-02-02 07:12:05 -05:00
Billy D.	c26e4e5ef0	feat: Add Kubeflow Pipeline definitions - voice_pipeline: STT → RAG → LLM → TTS - document_ingestion_pipeline: Extract → Chunk → Embed → Milvus - document_ingestion_mlflow_pipeline: With MLflow tracking - evaluation_pipeline: Model benchmarking - kfp-sync-job: K8s job to sync pipelines	2026-02-01 20:41:13 -05:00
Billy Davies	c36655b570	Initial commit	2026-02-02 01:40:30 +00:00

14 Commits