allenai--olmocr/scripts/run_chandra_benchmark.sh

#!/bin/bash

# Runs chandra benchmark, measuring both olmOCR-bench performance and per document processing performance
# Usage:
#   ./scripts/run_chandra_benchmark.sh                                      # Use latest chandra version, default benchmark repo
#   ./scripts/run_chandra_benchmark.sh 0.1.0                                # Use specific chandra version
#   ./scripts/run_chandra_benchmark.sh --version 0.1.0                      # Use specific chandra version (explicit flag)
#   ./scripts/run_chandra_benchmark.sh --benchrepo allenai/olmOCR-bench-internal  # Use different benchmark repo
#   ./scripts/run_chandra_benchmark.sh --benchbranch olmOCR-bench-1125      # Use specific branch/revision
#   ./scripts/run_chandra_benchmark.sh --benchpath s3://ai2-oe-data/path/   # Use benchmark from S3 or local path
#   ./scripts/run_chandra_benchmark.sh --cluster ai2/titan-cirrascale       # Specify a cluster
#   ./scripts/run_chandra_benchmark.sh --beaker-image jakep/olmocr-benchmark-0.3.3-780bc7d934  # Skip Docker build
#   ./scripts/run_chandra_benchmark.sh --noperf                             # Skip the performance test job
#   ./scripts/run_chandra_benchmark.sh --model datalab-to/chandra-ocr-2     # Use a specific model

set -e

# Parse command line arguments
CHANDRA_VERSION=""
BENCH_BRANCH=""
BENCH_REPO=""
BENCH_PATH=""
CLUSTER=""
BEAKER_IMAGE=""
NOPERF=""
MODEL=""

# Default to latest if no version specified
while [[ $# -gt 0 ]]; do
    case $1 in
        --version)
            CHANDRA_VERSION="$2"
            shift 2
            ;;
        --benchbranch)
            BENCH_BRANCH="$2"
            shift 2
            ;;
        --benchrepo)
            BENCH_REPO="$2"
            shift 2
            ;;
        --benchpath)
            BENCH_PATH="$2"
            shift 2
            ;;
        --cluster)
            CLUSTER="$2"
            shift 2
            ;;
        --beaker-image)
            BEAKER_IMAGE="$2"
            shift 2
            ;;
        --noperf)
            NOPERF="1"
            shift
            ;;
        --model)
            MODEL="$2"
            shift 2
            ;;
        *)
            # If no flag, assume it's the version for backward compatibility
            if [ -z "$CHANDRA_VERSION" ]; then
                CHANDRA_VERSION="$1"
            else
                echo "Unknown option: $1"
                echo "Usage: $0 [VERSION] [--version VERSION] [--benchbranch BRANCH] [--benchrepo REPO] [--benchpath PATH] [--cluster CLUSTER] [--beaker-image IMAGE] [--noperf] [--model MODEL]"
                exit 1
            fi
            shift
            ;;
    esac
done

# Set default chandra version if not specified
if [ -z "$CHANDRA_VERSION" ]; then
    CHANDRA_VERSION="latest"
fi

if [ "$CHANDRA_VERSION" = "latest" ]; then
    echo "Using latest chandra-ocr release"
    CHANDRA_INSTALL_CMD="uv pip install --system chandra-ocr"
else
    echo "Using chandra-ocr version: $CHANDRA_VERSION"
    CHANDRA_INSTALL_CMD="uv pip install --system chandra-ocr==$CHANDRA_VERSION"
fi

# Check for mutual exclusivity between benchpath and benchrepo/benchbranch
if [ -n "$BENCH_PATH" ] && ([ -n "$BENCH_REPO" ] || [ -n "$BENCH_BRANCH" ]); then
    echo "Error: --benchpath is mutually exclusive with --benchrepo and --benchbranch"
    echo "Use either --benchpath OR --benchrepo/--benchbranch, not both."
    exit 1
fi

# Use conda environment Python if available, otherwise use system Python
if [ -n "$CONDA_PREFIX" ]; then
    PYTHON="$CONDA_PREFIX/bin/python"
    echo "Using conda Python from: $CONDA_PREFIX"
else
    PYTHON="python"
    echo "Warning: No conda environment detected, using system Python"
fi

# Get version from version.py
VERSION=$($PYTHON -c 'import olmocr.version; print(olmocr.version.VERSION)')
echo "OlmOCR version: $VERSION"

# Get first 10 characters of git hash
GIT_HASH=$(git rev-parse HEAD | cut -c1-10)
echo "Git hash: $GIT_HASH"

# Get current git branch name
GIT_BRANCH=$(git rev-parse --abbrev-ref HEAD)
echo "Git branch: $GIT_BRANCH"

# Check if a Beaker image was provided
if [ -n "$BEAKER_IMAGE" ]; then
    echo "Using provided Beaker image: $BEAKER_IMAGE"
    IMAGE_TAG="$BEAKER_IMAGE"
else
    # Create full image tag
    IMAGE_TAG="olmocr-benchmark-${VERSION}-${GIT_HASH}"
    echo "Building Docker image with tag: $IMAGE_TAG"

    # Build the Docker image
    echo "Building Docker image..."
    docker build --platform linux/amd64 -f ./Dockerfile -t $IMAGE_TAG .

    # Push image to beaker
    echo "Trying to push image to Beaker..."
    if ! beaker image create --workspace ai2/oe-data-pdf --name $IMAGE_TAG $IMAGE_TAG 2>/dev/null; then
        echo "Warning: Beaker image with tag $IMAGE_TAG already exists. Using existing image."
    fi
fi

# Get Beaker username
BEAKER_USER=$(beaker account whoami --format json | jq -r '.[0].name')
echo "Beaker user: $BEAKER_USER"

# Create Python script to run beaker experiment
cat << 'EOF' > /tmp/run_benchmark_experiment.py
import sys
from textwrap import dedent
from beaker import Beaker, BeakerExperimentSpec, BeakerTaskSpec, BeakerTaskContext, BeakerResultSpec, BeakerTaskResources, BeakerImageSource, BeakerJobPriority, BeakerConstraints, BeakerEnvVar

# Get image tag, beaker user, git branch, git hash, and chandra version from command line
image_tag = sys.argv[1]
beaker_user = sys.argv[2]
git_branch = sys.argv[3]
git_hash = sys.argv[4]
chandra_version = sys.argv[5]
chandra_install_cmd = sys.argv[6]

# Initialize benchmark dataset parameters
bench_branch = None
bench_repo = "allenai/olmOCR-bench"  # Default repository
bench_path = None
cluster = None
noperf = False
model = None

# Parse additional arguments
arg_idx = 7
while arg_idx < len(sys.argv):
    if sys.argv[arg_idx] == "--benchbranch":
        bench_branch = sys.argv[arg_idx + 1]
        arg_idx += 2
    elif sys.argv[arg_idx] == "--benchrepo":
        bench_repo = sys.argv[arg_idx + 1]
        arg_idx += 2
    elif sys.argv[arg_idx] == "--benchpath":
        bench_path = sys.argv[arg_idx + 1]
        arg_idx += 2
    elif sys.argv[arg_idx] == "--cluster":
        cluster = sys.argv[arg_idx + 1]
        arg_idx += 2
    elif sys.argv[arg_idx] == "--noperf":
        noperf = True
        arg_idx += 1
    elif sys.argv[arg_idx] == "--model":
        model = sys.argv[arg_idx + 1]
        arg_idx += 2
    else:
        print(f"Unknown argument: {sys.argv[arg_idx]}")
        arg_idx += 1

# Default model for chandra
chandra_model = model if model else "datalab-to/chandra-ocr-2"

# Initialize Beaker client
b = Beaker.from_env(default_workspace="ai2/olmocr")

# Check if AWS credentials secret exists
aws_creds_secret = f"{beaker_user}-AWS_CREDENTIALS_FILE"
try:
    # Try to get the secret to see if it exists
    b.secret.get(aws_creds_secret, workspace="ai2/olmocr")
    has_aws_creds = True
    print(f"Found AWS credentials secret: {aws_creds_secret}")
except:
    has_aws_creds = False
    print(f"AWS credentials secret not found: {aws_creds_secret}")

# Shell script to run Chandra conversions for benchmark
run_chandra_shell = dedent("""\
bash -lc 'set -euo pipefail
PDF_ROOT="olmOCR-bench/bench_data/pdfs"
TARGET_ROOT="olmOCR-bench/bench_data/chandra"
rm -rf "$TARGET_ROOT"
mkdir -p "$TARGET_ROOT"

# Start vllm server in background
echo "Starting vllm server for Chandra..."
vllm serve __CHANDRA_MODEL__ --served-model-name chandra > /tmp/vllm_server.log 2>&1 &
VLLM_PID=$!

# Wait for vllm server to be ready
echo "Waiting for vllm server to start..."
for i in {1..600}; do
    if curl -s http://localhost:8000/health > /dev/null 2>&1; then
        echo "vllm server is ready"
        break
    fi
    if [ $i -eq 600 ]; then
        echo "Error: vllm server failed to start after 600 seconds"
        cat /tmp/vllm_server.log
        exit 1
    fi
    sleep 1
done

# Process each folder separately to maintain structure
echo "Running Chandra conversions..."
for folder in "$PDF_ROOT"/*; do
    if [ ! -d "$folder" ]; then
        continue
    fi
    section=$(basename "$folder")
    output_dir="$HOME/chandra_bench_${section}"
    rm -rf "$output_dir"
    mkdir -p "$output_dir"

    echo "  Processing $folder -> $output_dir"
    chandra "$folder" "$output_dir" --method vllm
done

echo "Collecting Chandra markdown outputs..."
# For each PDF, find its corresponding markdown and copy to proper location
find "$PDF_ROOT" -type f -name "*.pdf" | while IFS= read -r pdf_path; do
    # Get relative path from PDF root
    rel_path=${pdf_path#"$PDF_ROOT"/}

    # Extract section name (first directory level)
    case "$rel_path" in
        */*)
            section=${rel_path%%/*}
            ;;
        *)
            echo "Warning: Unexpected PDF path layout for $pdf_path, skipping" >&2
            continue
            ;;
    esac

    # Get PDF name without extension
    pdf_name=$(basename "$pdf_path" .pdf)

    # Source markdown location: ~/chandra_bench_${section}/${pdf_name}/${pdf_name}.md
    src_md="$HOME/chandra_bench_${section}/${pdf_name}/${pdf_name}.md"

    if [ ! -f "$src_md" ]; then
        echo "Warning: No markdown output found at $src_md" >&2
        continue
    fi

    # Target location: olmOCR-bench/bench_data/chandra/${section}/${pdf_name}_pg1_repeat1.md
    target_dir="$TARGET_ROOT/$section"
    mkdir -p "$target_dir"
    target_path="$target_dir/${pdf_name}_pg1_repeat1.md"

    cp "$src_md" "$target_path"
    echo "  Copied $src_md -> $target_path"
done

# Kill vllm server
echo "Stopping vllm server..."
kill $VLLM_PID || true
wait $VLLM_PID 2>/dev/null || true
'""").replace("__CHANDRA_MODEL__", chandra_model)

# Check if HF_TOKEN secret exists
hf_token_secret = f"{beaker_user}-HF_TOKEN"
try:
    # Try to get the secret to see if it exists
    b.secret.get(hf_token_secret, workspace="ai2/olmocr")
    has_hf_token = True
    print(f"Found HuggingFace token secret: {hf_token_secret}")
except:
    has_hf_token = False
    print(f"HuggingFace token secret not found: {hf_token_secret}")

# First experiment: Original benchmark job
commands = []
if has_aws_creds:
    commands.extend([
        "mkdir -p ~/.aws",
        'echo "$AWS_CREDENTIALS_FILE" > ~/.aws/credentials'
    ])

if has_hf_token:
    commands.append('export HF_TOKEN="$HF_TOKEN"')

# Install uv for fast dependency management, then s5cmd (needed for S3 operations)
commands.append("pip install uv")
commands.append("uv pip install --system s5cmd")

# Handle benchmark data download based on source type
if bench_path:
    # If bench_path is provided, use it (can be S3 or local path)
    if bench_path.startswith("s3://"):
        # S3 path - use s5cmd to download
        commands.append(f"s5cmd cp {bench_path.rstrip('/')}/* ./olmOCR-bench/")
    else:
        # Local path - copy directly
        commands.append(f"cp -r {bench_path} ./olmOCR-bench")
else:
    # Use HuggingFace download (default behavior)
    hf_download_cmd = f"hf download --repo-type dataset {bench_repo} --max-workers 2"
    if bench_branch:
        hf_download_cmd += f" --revision {bench_branch}"
    hf_download_cmd += " --local-dir ./olmOCR-bench"
    commands.append(hf_download_cmd)

commands.extend([
    chandra_install_cmd,
    "uv pip uninstall --system vllm flashinfer flashinfer-python flashinfer-cubin || true",
    "uv pip install --system vllm --torch-backend=auto",
    run_chandra_shell,
    "python -m olmocr.bench.benchmark --dir ./olmOCR-bench/bench_data --candidate chandra"
])

# Build task spec with optional env vars
# If image_tag contains '/', it's already a full beaker image reference
if '/' in image_tag:
    image_ref = image_tag
else:
    image_ref = f"{beaker_user}/{image_tag}"

task_spec_args = {
    "name": "chandra-benchmark",
    "image": BeakerImageSource(beaker=image_ref),
    "command": [
        "bash", "-c",
        " && ".join(commands)
    ],
    "context": BeakerTaskContext(
        priority=BeakerJobPriority["normal"],
        preemptible=True,
    ),
    "resources": BeakerTaskResources(gpu_count=1),
    "constraints": BeakerConstraints(cluster=[cluster] if cluster else ["ai2/ceres-cirrascale", "ai2/jupiter-cirrascale-2"]),
    "result": BeakerResultSpec(path="/noop-results"),
}

# Add env vars if AWS credentials or HF token exist
env_vars = []
if has_aws_creds:
    env_vars.append(BeakerEnvVar(name="AWS_CREDENTIALS_FILE", secret=aws_creds_secret))
if has_hf_token:
    env_vars.append(BeakerEnvVar(name="HF_TOKEN", secret=hf_token_secret))
if env_vars:
    task_spec_args["env_vars"] = env_vars

# Create first experiment spec
chandra_version_label = "latest" if chandra_version == "latest" else chandra_version
experiment_spec = BeakerExperimentSpec(
    description=f"Chandra {chandra_version_label} Benchmark Run - Branch: {git_branch}, Commit: {git_hash}",
    budget="ai2/oe-base",
    tasks=[BeakerTaskSpec(**task_spec_args)],
)

# Create the first experiment
workload = b.experiment.create(spec=experiment_spec, workspace="ai2/olmocr")
print(f"Created benchmark experiment: {workload.experiment.id}")
print(f"View at: https://beaker.org/ex/{workload.experiment.id}")
print("-------")
print("")

# Second experiment: Performance test job (only if --noperf not specified)
if not noperf:
    perf_commands = []
    if has_aws_creds:
        perf_commands.extend([
            "mkdir -p ~/.aws",
            'echo "$AWS_CREDENTIALS_FILE" > ~/.aws/credentials'
        ])

    if has_hf_token:
        perf_commands.append('export HF_TOKEN="$HF_TOKEN"')

    # Shell script for performance test
    perf_shell = dedent("""\
set -euo pipefail

# Start vllm server in background
echo "Starting vllm server for Chandra..."
vllm serve __CHANDRA_MODEL__ --served-model-name chandra > /tmp/vllm_server.log 2>&1 &
VLLM_PID=$!

# Wait for vllm server to be ready
echo "Waiting for vllm server to start..."
for i in {1..600}; do
    if curl -s http://localhost:8000/health > /dev/null 2>&1; then
        echo "vllm server is ready"
        break
    fi
    if [ $i -eq 600 ]; then
        echo "Error: vllm server failed to start after 600 seconds"
        cat /tmp/vllm_server.log
        exit 1
    fi
    sleep 1
done

# Run the performance test
time chandra /root/olmOCR-mix-0225_benchmark_set/ /root/olmOCR-mix-0225_benchmark_set_chandra --method vllm

# Kill vllm server
echo "Stopping vllm server..."
kill $VLLM_PID || true
wait $VLLM_PID 2>/dev/null || true
""").replace("__CHANDRA_MODEL__", chandra_model)

    perf_commands.extend([
        "pip install uv",
        chandra_install_cmd,
        "uv pip uninstall --system vllm flashinfer flashinfer-python flashinfer-cubin || true",
        "uv pip install --system vllm --torch-backend=auto",
        "uv pip install --system awscli",
        "aws s3 cp --recursive s3://ai2-oe-data/jakep/olmocr/olmOCR-mix-0225/benchmark_set/ /root/olmOCR-mix-0225_benchmark_set/",
        f"bash -c '{perf_shell}'"
    ])

    # Build performance task spec
    perf_task_spec_args = {
        "name": "chandra-performance",
        "image": BeakerImageSource(beaker=image_ref),
        "command": [
            "bash", "-c",
            " && ".join(perf_commands)
        ],
        "context": BeakerTaskContext(
            priority=BeakerJobPriority["normal"],
            preemptible=True,
        ),
        # Need to reserve all 8 gpus for performance spec or else benchmark results can be off (1 for titan-cirrascale)
        "resources": BeakerTaskResources(gpu_count=1 if cluster == "ai2/titan-cirrascale" else 8),
        "constraints": BeakerConstraints(cluster=[cluster] if cluster else ["ai2/ceres-cirrascale", "ai2/jupiter-cirrascale-2"]),
        "result": BeakerResultSpec(path="/noop-results"),
    }

    # Add env vars if AWS credentials or HF token exist
    env_vars = []
    if has_aws_creds:
        env_vars.append(BeakerEnvVar(name="AWS_CREDENTIALS_FILE", secret=aws_creds_secret))
    if has_hf_token:
        env_vars.append(BeakerEnvVar(name="HF_TOKEN", secret=hf_token_secret))
    if env_vars:
        perf_task_spec_args["env_vars"] = env_vars

    # Create performance experiment spec
    perf_experiment_spec = BeakerExperimentSpec(
        description=f"Chandra {chandra_version_label} Performance Test - Branch: {git_branch}, Commit: {git_hash}",
        budget="ai2/oe-base",
        tasks=[BeakerTaskSpec(**perf_task_spec_args)],
    )

    # Create the performance experiment
    perf_workload = b.experiment.create(spec=perf_experiment_spec, workspace="ai2/olmocr")
    print(f"Created performance experiment: {perf_workload.experiment.id}")
    print(f"View at: https://beaker.org/ex/{perf_workload.experiment.id}")
else:
    print("Skipping performance test (--noperf flag specified)")
EOF

# Run the Python script to create the experiments
echo "Creating Beaker experiments..."

# Build command with appropriate arguments
CMD="$PYTHON /tmp/run_benchmark_experiment.py $IMAGE_TAG $BEAKER_USER $GIT_BRANCH $GIT_HASH \"$CHANDRA_VERSION\" \"$CHANDRA_INSTALL_CMD\""

if [ -n "$BENCH_BRANCH" ]; then
    echo "Using bench branch: $BENCH_BRANCH"
    CMD="$CMD --benchbranch \"$BENCH_BRANCH\""
fi

if [ -n "$BENCH_REPO" ]; then
    echo "Using bench repo: $BENCH_REPO"
    CMD="$CMD --benchrepo \"$BENCH_REPO\""
fi

if [ -n "$BENCH_PATH" ]; then
    echo "Using bench path: $BENCH_PATH"
    CMD="$CMD --benchpath \"$BENCH_PATH\""
fi

if [ -n "$CLUSTER" ]; then
    echo "Using cluster: $CLUSTER"
    CMD="$CMD --cluster $CLUSTER"
fi

if [ -n "$NOPERF" ]; then
    echo "Skipping performance tests"
    CMD="$CMD --noperf"
fi

if [ -n "$MODEL" ]; then
    echo "Using model: $MODEL"
    CMD="$CMD --model $MODEL"
fi

eval $CMD

# Clean up temporary file
rm /tmp/run_benchmark_experiment.py

echo "Benchmark experiments submitted successfully!"