Initial snapshot before transformerlab recovery

2026-03-31 13:25:25 -06:00
commit d860318d43
49 changed files with 3864 additions and 0 deletions
@@ -0,0 +1,170 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+SCRIPT_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
+STATE_DIR=$(cd "$SCRIPT_DIR/.." && pwd)
+RUNTIME_ENV="$STATE_DIR/runtime.env"
+
+if [ ! -f "$RUNTIME_ENV" ]; then
+  echo "Missing $RUNTIME_ENV. Run ./labctl up first." >&2
+  exit 1
+fi
+
+# shellcheck disable=SC1090
+. "$RUNTIME_ENV"
+
+REPO_URL="https://huggingface.co/{{ courseware_white_rabbit_repo }}"
+REPO_REF="main"
+REPO_DIR="$COURSEWARE_STATE_DIR/downloads/WhiteRabbitNeo_WhiteRabbitNeo-V3-7B-GGUF"
+TARGET_DIR="$COURSEWARE_STATE_DIR/models/WhiteRabbitNeo"
+REGISTER_WITH_OLLAMA=1
+
+DOWNLOAD_FILES=(
+{% for variant in courseware_white_rabbit_variants | unique(attribute='filename') %}
+  "{{ variant.filename }}"
+{% endfor %}
+)
+
+MODEL_NAMES=(
+{% for variant in courseware_white_rabbit_variants %}
+  "{{ variant.ollama_model }}"
+{% endfor %}
+)
+
+MODEL_FILES=(
+{% for variant in courseware_white_rabbit_variants %}
+  "{{ variant.filename }}"
+{% endfor %}
+)
+
+usage() {
+  cat <<'EOF'
+Usage: ./download_whiterabbitneo-gguf.sh [--download-only]
+
+Downloads the WhiteRabbitNeo GGUF variants used in lab 2 with git + git-lfs.
+By default it also registers local Ollama aliases after the files are present.
+
+Options:
+  --download-only   Skip Ollama model registration
+  -h, --help        Show this help text
+EOF
+}
+
+require_cmd() {
+  if ! command -v "$1" >/dev/null 2>&1; then
+    echo "Missing required command: $1" >&2
+    exit 1
+  fi
+}
+
+parse_args() {
+  while [ $# -gt 0 ]; do
+    case "$1" in
+      --download-only)
+        REGISTER_WITH_OLLAMA=0
+        ;;
+      -h|--help)
+        usage
+        exit 0
+        ;;
+      *)
+        echo "Unknown option: $1" >&2
+        usage >&2
+        exit 1
+        ;;
+    esac
+    shift
+  done
+}
+
+setup_repo() {
+  mkdir -p "$COURSEWARE_STATE_DIR/downloads"
+
+  if [ ! -d "$REPO_DIR/.git" ]; then
+    rm -rf "$REPO_DIR"
+    GIT_LFS_SKIP_SMUDGE=1 git clone --filter=blob:none --no-checkout "$REPO_URL" "$REPO_DIR"
+  fi
+
+  git -C "$REPO_DIR" remote set-url origin "$REPO_URL"
+  git -C "$REPO_DIR" sparse-checkout init --no-cone
+  git -C "$REPO_DIR" sparse-checkout set -- "${DOWNLOAD_FILES[@]}"
+  git -C "$REPO_DIR" fetch --depth=1 origin "$REPO_REF"
+  git -C "$REPO_DIR" checkout -f --detach FETCH_HEAD
+  git -C "$REPO_DIR" lfs install --local >/dev/null
+}
+
+download_files() {
+  local lfs_include
+
+  mkdir -p "$TARGET_DIR"
+  lfs_include=$(IFS=,; printf '%s' "${DOWNLOAD_FILES[*]}")
+
+  git -C "$REPO_DIR" lfs pull origin --include="$lfs_include" --exclude=""
+
+  for file in "${DOWNLOAD_FILES[@]}"; do
+    cp -f "$REPO_DIR/$file" "$TARGET_DIR/$file"
+  done
+}
+
+start_ollama_if_needed() {
+  if curl -fsS "http://$COURSEWARE_BIND_HOST:$COURSEWARE_OLLAMA_PORT/api/tags" >/dev/null 2>&1; then
+    return 0
+  fi
+
+  if [ -x "$COURSEWARE_ROOT/scripts/service_manager.sh" ]; then
+    "$COURSEWARE_ROOT/scripts/service_manager.sh" start ollama >/dev/null
+  fi
+
+  curl -fsS "http://$COURSEWARE_BIND_HOST:$COURSEWARE_OLLAMA_PORT/api/tags" >/dev/null 2>&1
+}
+
+register_models() {
+  local i
+  local modelfile
+
+  if [ "$REGISTER_WITH_OLLAMA" -eq 0 ]; then
+    echo "Skipping Ollama registration because --download-only was requested."
+    return 0
+  fi
+
+  if ! command -v "$OLLAMA_BIN" >/dev/null 2>&1; then
+    echo "Ollama is not installed or not on PATH; downloads completed but model registration was skipped." >&2
+    return 0
+  fi
+
+  if ! start_ollama_if_needed; then
+    echo "Ollama is not reachable on http://$COURSEWARE_BIND_HOST:$COURSEWARE_OLLAMA_PORT; downloads completed but model registration was skipped." >&2
+    return 0
+  fi
+
+  export OLLAMA_HOST="$COURSEWARE_BIND_HOST:$COURSEWARE_OLLAMA_PORT"
+  export OLLAMA_MODELS="$OLLAMA_MODELS_DIR"
+
+  for i in "${!MODEL_NAMES[@]}"; do
+    modelfile="$TARGET_DIR/Modelfile.${MODEL_NAMES[$i]}"
+    printf 'FROM %s/%s\n' "$TARGET_DIR" "${MODEL_FILES[$i]}" >"$modelfile"
+    "$OLLAMA_BIN" create "${MODEL_NAMES[$i]}" -f "$modelfile"
+  done
+}
+
+main() {
+  parse_args "$@"
+  require_cmd git
+  require_cmd git-lfs
+  git lfs install --skip-repo >/dev/null
+
+  echo "Preparing WhiteRabbitNeo GGUF checkout in $REPO_DIR"
+  setup_repo
+
+  echo "Downloading selected WhiteRabbitNeo GGUF variants into $TARGET_DIR"
+  download_files
+
+  echo "Download complete:"
+  printf '  - %s\n' "${DOWNLOAD_FILES[@]}"
+
+  register_models
+
+  echo "WhiteRabbitNeo lab 2 setup is ready."
+}
+
+main "$@"
@@ -0,0 +1,14 @@
+description: Local evaluation placeholder for lab 6
+providers:
+  - id: openai:chat
+    label: Local Ollama
+    config:
+      apiBaseUrl: http://{{ courseware_bind_host }}:{{ courseware_ports.ollama }}/v1
+      apiKey: local-only
+      model: qwen3.5:4b
+prompts:
+  - "{{ '{{prompt}}' }}"
+tests:
+  - vars:
+      prompt: "Summarize the purpose of this lab environment in one sentence."
+
@@ -0,0 +1,30 @@
+COURSEWARE_ROOT="{{ courseware_root }}"
+COURSEWARE_STATE_DIR="{{ courseware_state_dir }}"
+COURSEWARE_BIND_HOST="{{ courseware_bind_host }}"
+COURSEWARE_URL_HOST="{{ courseware_url_host }}"
+COURSEWARE_OLLAMA_PORT="{{ courseware_ports.ollama }}"
+COURSEWARE_OPEN_WEBUI_PORT="{{ courseware_ports.open_webui }}"
+COURSEWARE_TRANSFORMERLAB_PORT="{{ courseware_ports.transformerlab }}"
+COURSEWARE_CHUNKVIZ_PORT="{{ courseware_ports.chunkviz }}"
+COURSEWARE_EMBEDDING_ATLAS_PORT="{{ courseware_ports.embedding_atlas }}"
+COURSEWARE_UNSLOTH_PORT="{{ courseware_ports.unsloth }}"
+COURSEWARE_PROMPTFOO_PORT="{{ courseware_ports.promptfoo }}"
+COURSEWARE_WIKI_PORT="{{ courseware_ports.wiki }}"
+OLLAMA_BIN="{{ courseware_ollama_bin }}"
+OLLAMA_MODELS_DIR="{{ courseware_ollama_models_dir }}"
+NODE_RUNTIME_BIN_DIR="{{ courseware_node_runtime_bin_dir }}"
+OPEN_WEBUI_VENV="{{ courseware_venvs_dir }}/open-webui"
+OPEN_WEBUI_DATA_DIR="{{ courseware_state_dir }}/open-webui"
+CHUNKVIZ_DIR="{{ courseware_repos_dir }}/ChunkViz"
+EMBEDDING_ATLAS_VENV="{{ courseware_venvs_dir }}/embedding-atlas"
+TTPS_DATASET_PATH="{{ courseware_datasets_dir }}/ttps_dataset.parquet"
+WIKI_TEST_RAW_PATH="{{ courseware_datasets_dir }}/wiki.test.raw"
+TRANSFORMERLAB_DIR="{{ courseware_transformerlab_home }}"
+UNSLOTH_BIN="{{ ansible_env.HOME }}/.local/bin/unsloth"
+PROMPTFOO_DIR="{{ courseware_promptfoo_dir }}"
+PROMPTFOO_BIN="{{ courseware_tools_dir }}/promptfoo/node_modules/.bin/promptfoo"
+WIKI_DIR="{{ courseware_wiki_repo_dir }}"
+LLAMA_CPP_BIN_DIR="{{ courseware_llama_cpp_bin_dir }}"
+KILN_LINUX_BIN="{{ courseware_apps_dir }}/kiln/Kiln"
+KILN_MAC_APP="{{ courseware_apps_dir }}/Kiln.app"
+KILN_LAUNCH_PATH="{% if ansible_system == 'Darwin' %}{{ courseware_apps_dir }}/Kiln.app{% else %}{{ courseware_apps_dir }}/kiln/Kiln{% endif %}"