upload.sh

#!/bin/bash
set -euo pipefail

# =============================================================================
# TinyCLAP CoreML — HuggingFace Upload Script
# =============================================================================
# Usage:
#   export HF_TOKEN=hf_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
#   ./upload.sh [org/username]
#
# The optional argument is the HF namespace. Defaults to the current user's
# namespace if omitted.
# =============================================================================

REPO_NAME="tinyclap-coreml"
NAMESPACE="${1:-}"

if [ -z "$NAMESPACE" ]; then
    echo "Usage: $0 <hf-namespace>"
    echo "Example: $0 duhamelc"
    exit 1
fi

FULL_REPO="$NAMESPACE/$REPO_NAME"

echo "========================================"
echo "  TinyCLAP CoreML HF Upload"
echo "  Target: https://huggingface.co/$FULL_REPO"
echo "========================================"
echo

# --- Check prerequisites ------------------------------------------------------

if [ -z "${HF_TOKEN:-}" ]; then
    echo "Error: HF_TOKEN environment variable is not set."
    echo "Get your token from https://huggingface.co/settings/tokens"
    exit 1
fi

if ! command -v git &> /dev/null; then
    echo "Error: git is not installed."
    exit 1
fi

if ! command -v git-lfs &> /dev/null; then
    echo "Error: git-lfs is not installed. Install it with: brew install git-lfs"
    exit 1
fi

if ! command -v huggingface-cli &> /dev/null; then
    echo "Warning: huggingface-cli not found. Will use raw git + LFS instead."
    USE_CLI=false
else
    USE_CLI=true
fi

# --- Verify files exist -------------------------------------------------------

FILES=(
    "TinyCLAP_AudioEncoder.mlpackage"
    "TinyCLAP_TextEncoder.mlpackage"
    "mel_filter_bank.json"
    "text_embeddings.json"
    "README.md"
    "LICENSE"
    ".gitattributes"
)

for f in "${FILES[@]}"; do
    if [ ! -e "$f" ]; then
        echo "Error: Required file '$f' not found in current directory."
        exit 1
    fi
done

# --- Initialize git repo if needed --------------------------------------------

if [ ! -d ".git" ]; then
    echo "→ Initializing git repository..."
    git init
    git lfs install
fi

# --- Create or use existing remote -------------------------------------------

if ! git remote get-url origin &> /dev/null; then
    echo "→ Adding remote origin..."
    git remote add origin "https://huggingface.co/$FULL_REPO"
else
    echo "→ Updating remote origin..."
    git remote set-url origin "https://huggingface.co/$FULL_REPO"
fi

# --- Stage and commit ---------------------------------------------------------

echo "→ Staging files..."
git add -A

if git diff --cached --quiet; then
    echo "Nothing to commit. Repository is up to date."
else
    echo "→ Committing..."
    git commit --author="Velvox Builder <build@localhost>" -m "Update TinyCLAP CoreML models — ANE/GPU enabled

Changes:
- Recompiled with compute_units=ALL (CPU + GPU + Apple Neural Engine)
- Eliminates CPU-only fallback during inference
- Numerically verified: audio cos_sim ≈ 1.0, text cos_sim ≈ 0.999996

Files:
- TinyCLAP_AudioEncoder.mlmodel (~17 MB)
- TinyCLAP_TextEncoder.mlmodel (~422 MB)
- mel_filter_bank.json"
fi

# --- Push ---------------------------------------------------------------------

echo "→ Pushing to HuggingFace..."

if [ "$USE_CLI" = true ]; then
    huggingface-cli upload "$FULL_REPO" . . --repo-type model --token "$HF_TOKEN"
else
    # Raw git push with token in URL
    git push "https://user:$HF_TOKEN@huggingface.co/$FULL_REPO" main --force
fi

echo
echo "========================================"
echo "  Upload complete!"
echo "  https://huggingface.co/$FULL_REPO"
echo "========================================"