Rebase may 07

.buildkite/lm-eval-harness/conftest.py

@@ -0,0 +1,39 @@
+# SPDX-License-Identifier: Apache-2.0
+from pathlib import Path
+
+import pytest
+
+
+def pytest_addoption(parser):
+    parser.addoption(
+        "--config-list-file",
+        action="store",
+        help="Path to the file listing model config YAMLs (one per line)")
+    parser.addoption("--tp-size",
+                     action="store",
+                     default="1",
+                     help="Tensor parallel size to use for evaluation")
+
+
+@pytest.fixture(scope="session")
+def config_list_file(pytestconfig, config_dir):
+    rel_path = pytestconfig.getoption("--config-list-file")
+    return config_dir / rel_path
+
+
+@pytest.fixture(scope="session")
+def tp_size(pytestconfig):
+    return pytestconfig.getoption("--tp-size")
+
+
+def pytest_generate_tests(metafunc):
+    if "config_filename" in metafunc.fixturenames:
+        rel_path = metafunc.config.getoption("--config-list-file")
+        config_list_file = Path(rel_path).resolve()
+        config_dir = config_list_file.parent
+        with open(config_list_file, encoding="utf-8") as f:
+            configs = [
+                config_dir / line.strip() for line in f
+                if line.strip() and not line.startswith("#")
+            ]
+        metafunc.parametrize("config_filename", configs)

.buildkite/lm-eval-harness/test_lm_eval_correctness.py

@@ -3,67 +3,48 @@
 LM eval harness on model to compare vs HF baseline computed offline.
 Configs are found in configs/$MODEL.yaml
 
-* export LM_EVAL_TEST_DATA_FILE=configs/Meta-Llama-3-70B-Instruct.yaml
-* export LM_EVAL_TP_SIZE=4 
-* pytest -s test_lm_eval_correctness.py
+pytest -s -v test_lm_eval_correctness.py \
+    --config-list-file=configs/models-small.txt \
+    --tp-size=1
 """
 
-import os
-from pathlib import Path
-
 import lm_eval
-import numpy
-import pytest
+import numpy as np
 import yaml
 
 RTOL = 0.08
-TEST_DATA_FILE = os.environ.get(
-    "LM_EVAL_TEST_DATA_FILE",
-    ".buildkite/lm-eval-harness/configs/Meta-Llama-3-8B-Instruct.yaml")
-
-TP_SIZE = os.environ.get("LM_EVAL_TP_SIZE", 1)
 
 
-def launch_lm_eval(eval_config):
+def launch_lm_eval(eval_config, tp_size):
     trust_remote_code = eval_config.get('trust_remote_code', False)
-
     model_args = f"pretrained={eval_config['model_name']}," \
-                 f"tensor_parallel_size={TP_SIZE}," \
+                 f"tensor_parallel_size={tp_size}," \
+                 f"enforce_eager=true," \
                  f"add_bos_token=true," \
                  f"trust_remote_code={trust_remote_code}"
-
     results = lm_eval.simple_evaluate(
         model="vllm",
         model_args=model_args,
         tasks=[task["name"] for task in eval_config["tasks"]],
         num_fewshot=eval_config["num_fewshot"],
         limit=eval_config["limit"],
         batch_size="auto")
-
     return results
 
 
-def test_lm_eval_correctness():
-    eval_config = yaml.safe_load(
-        Path(TEST_DATA_FILE).read_text(encoding="utf-8"))
-
-    if eval_config[
-            "model_name"] == "nm-testing/Meta-Llama-3-70B-Instruct-FBGEMM-nonuniform":  #noqa: E501
-        pytest.skip("FBGEMM is currently failing on main.")
+def test_lm_eval_correctness_param(config_filename, tp_size):
+    eval_config = yaml.safe_load(config_filename.read_text(encoding="utf-8"))
 
-    # Launch eval requests.
-    results = launch_lm_eval(eval_config)
+    results = launch_lm_eval(eval_config, tp_size)
 
-    # Confirm scores match ground truth.
     success = True
     for task in eval_config["tasks"]:
         for metric in task["metrics"]:
             ground_truth = metric["value"]
             measured_value = results["results"][task["name"]][metric["name"]]
             print(f'{task["name"]} | {metric["name"]}: '
                   f'ground_truth={ground_truth} | measured={measured_value}')
-            success = success and numpy.isclose(
+            success = success and np.isclose(
                 ground_truth, measured_value, rtol=RTOL)
 
-    # Assert at the end, print all scores even on failure for debugging.
     assert success

.buildkite/scripts/hardware_ci/run-tpu-v1-test.sh

@@ -47,7 +47,9 @@ docker run --privileged --net host --shm-size=16G -it \
     && echo TEST_10 \
     && pytest -s -v /workspace/vllm/tests/v1/tpu/test_pallas.py \
     && echo TEST_11 \
-    && pytest -s -v /workspace/vllm/tests/v1/entrypoints/llm/test_struct_output_generate.py" \
+    && pytest -s -v /workspace/vllm/tests/v1/entrypoints/llm/test_struct_output_generate.py \
+    && echo TEST_12 \
+    && pytest -s -v /workspace/vllm/tests/tpu/test_moe_pallas.py" \
 
 
 # TODO: This test fails because it uses RANDOM_SEED sampling

.buildkite/test-pipeline.yaml

@@ -408,7 +408,7 @@ steps:
   - vllm/model_executor/layers/quantization
   commands:
   - export VLLM_WORKER_MULTIPROC_METHOD=spawn
-  - bash ./run-tests.sh -c configs/models-small.txt -t 1
+  - pytest -s -v test_lm_eval_correctness.py --config-list-file=configs/models-small.txt --tp-size=1
 
 - label: OpenAI API correctness
   source_file_dependencies:
@@ -713,4 +713,4 @@ steps:
   - vllm/model_executor/layers/quantization
   commands:
   - export VLLM_WORKER_MULTIPROC_METHOD=spawn
-  - bash ./run-tests.sh -c configs/models-large.txt -t 4
+  - pytest -s -v test_lm_eval_correctness.py --config-list-file=configs/models-large.txt --tp-size=4

benchmarks/benchmark_dataset.py

@@ -887,6 +887,94 @@ def sample(self,
         return sampled_requests
 
 
+# -----------------------------------------------------------------------------
+# Next Edit Prediction Dataset Implementation
+# -----------------------------------------------------------------------------
+
+
+zeta_prompt = """### Instruction:
+You are a code completion assistant and your task is to analyze user edits and then rewrite an excerpt that the user provides, suggesting the appropriate edits within the excerpt, taking into account the cursor location.
+
+### User Edits:
+
+{}
+
+### User Excerpt:
+
+{}
+
+### Response:
+
+""" # noqa: E501
+
+
+def _format_zeta_prompt(
+        sample: dict,
+        original_start_marker: str = "<|editable_region_start|>") -> dict:
+    """Format the zeta prompt for the Next Edit Prediction (NEP) dataset.
+
+    This function formats examples from the NEP dataset 
+    into prompts and expected outputs. It could be 
+    further extended to support more NEP datasets.
+
+    Args:
+        sample: The dataset sample containing events, 
+            inputs, and outputs.
+        original_start_marker: The marker indicating the 
+            start of the editable region. Defaults to 
+            "<|editable_region_start|>".
+
+    Returns:
+        A dictionary with the formatted prompts and expected outputs.
+    """
+    events = sample["events"]
+    input = sample["input"]
+    output = sample["output"]
+    prompt = zeta_prompt.format(events, input)
+
+    # following the original implementation, extract the focused region
+    # from the raw output
+    output_start_index = output.find(original_start_marker)
+    output_focused_region = output[output_start_index:]
+    expected_output = output_focused_region
+
+    return {"prompt": prompt, "expected_output": expected_output}
+
+
+class NextEditPredictionDataset(HuggingFaceDataset):
+    """
+    Dataset class for processing a Next Edit Prediction dataset.
+    """
+
+    SUPPORTED_DATASET_PATHS = {
+        "zed-industries/zeta",
+    }
+    MAPPING_PROMPT_FUNCS = {
+        "zed-industries/zeta": _format_zeta_prompt,
+    }
+
+    def sample(self, tokenizer: PreTrainedTokenizerBase, num_requests: int,
+               **kwargs):
+        formatting_prompt_func = self.MAPPING_PROMPT_FUNCS.get(
+            self.dataset_path)
+        if formatting_prompt_func is None:
+            raise ValueError(f"Unsupported dataset path: {self.dataset_path}")
+        samples = []
+        for sample in self.data:
+            sample = formatting_prompt_func(sample)
+            samples.append(
+                SampleRequest(
+                    prompt=sample["prompt"],
+                    prompt_len=len(tokenizer(sample["prompt"]).input_ids),
+                    expected_output_len=len(
+                        tokenizer(sample["expected_output"]).input_ids),
+                ))
+            if len(samples) >= num_requests:
+                break
+        self.maybe_oversample_requests(samples, num_requests)
+        return samples
+
+
 # -----------------------------------------------------------------------------
 # ASR Dataset Implementation
 # -----------------------------------------------------------------------------

benchmarks/benchmark_serving.py

@@ -53,8 +53,9 @@
 from benchmark_dataset import (AIMODataset, ASRDataset, BurstGPTDataset,
                                ConversationDataset, HuggingFaceDataset,
                                InstructCoderDataset, MTBenchDataset,
-                               RandomDataset, SampleRequest, ShareGPTDataset,
-                               SonnetDataset, VisionArenaDataset)
+                               NextEditPredictionDataset, RandomDataset,
+                               SampleRequest, ShareGPTDataset, SonnetDataset,
+                               VisionArenaDataset)
 from benchmark_utils import convert_to_pytorch_benchmark_format, write_to_json
 
 MILLISECONDS_TO_SECONDS_CONVERSION = 1000
@@ -603,6 +604,9 @@ def main(args: argparse.Namespace):
         elif args.dataset_path in AIMODataset.SUPPORTED_DATASET_PATHS:
             dataset_class = AIMODataset
             args.hf_split = "train"
+        elif args.dataset_path in NextEditPredictionDataset.SUPPORTED_DATASET_PATHS:  # noqa: E501
+            dataset_class = NextEditPredictionDataset
+            args.hf_split = "train"
         elif args.dataset_path in ASRDataset.SUPPORTED_DATASET_PATHS:
             dataset_class = ASRDataset
             args.hf_split = "train"