Move examples/hf_llm to examples/hf and share HF export reporting (pytorch#23898)

Summary:
Pull Request resolved: pytorch#23898

Rename `backends/native/examples/hf_llm` to `examples/hf` in preparation for non-LLM Hugging Face exporters, and factor delegated-versus-residual op reporting into `hf_utils.py` for reuse.

`export_hf_llm` and `convert_meta_llama` retain their flags and exported artifacts. Diagnostic report headers now include the method label, and an empty residual-op section prints `(none)`.

Updated with AI assistance from Codex.

Differential Revision: D125295022

Author: JakeStevens

backends/native/examples/hf_llm/BUCK backends/native/examples/hf/BUCKbackends/native/examples/hf_llm/BUCK renamed to backends/native/examples/hf/BUCK

@@ -3,12 +3,24 @@ load("@fbsource//xplat/executorch/build:runtime_wrapper.bzl", "runtime")
 
 oncall("executorch")
 
+fbcode_target(
+    _kind = runtime.python_library,
+    name = "hf_utils",
+    srcs = ["hf_utils.py"],
+    deps = [
+        "//caffe2:torch",
+        "//executorch/backends/native:lib",
+        "//executorch/exir:lib",
+    ],
+)
+
 fbcode_target(
     _kind = runtime.python_library,
     name = "export_hf_llm_lib",
     srcs = ["export_hf_llm.py"],
     deps = [
         ":gguf_hf",
+        ":hf_utils",
         "//caffe2:torch",
         "//executorch/backends/native:lib",
         "//executorch/exir:lib",
@@ -23,7 +35,7 @@ fbcode_target(
 fbcode_target(
     _kind = runtime.python_binary,
     name = "export_hf_llm",
-    main_function = "executorch.backends.native.examples.hf_llm.export_hf_llm.main",
+    main_function = "executorch.backends.native.examples.hf.export_hf_llm.main",
     deps = [":export_hf_llm_lib"],
 )
 
@@ -43,7 +55,7 @@ fbcode_target(
     _kind = runtime.python_binary,
     name = "convert_meta_llama",
     srcs = ["convert_meta_llama.py"],
-    main_function = "executorch.backends.native.examples.hf_llm.convert_meta_llama.main",
+    main_function = "executorch.backends.native.examples.hf.convert_meta_llama.main",
     deps = [
         "//caffe2:torch",
         "fbsource//third-party/pypi/safetensors:safetensors",

…ve/examples/hf_llm/convert_meta_llama.py …native/examples/hf/convert_meta_llama.pybackends/native/examples/hf_llm/convert_meta_llama.py renamed to backends/native/examples/hf/convert_meta_llama.py backends/native/examples/hf_llm/convert_meta_llama.py renamed to backends/native/examples/hf/convert_meta_llama.py

@@ -16,24 +16,12 @@
 """
 
 import argparse
-import collections
 import logging
-import operator
 
 import torch
 import torch.nn as nn
-from executorch.backends.native import get_default_compile_config
-from executorch.backends.native.examples.hf_llm.gguf_hf import (
-    gguf_architecture,
-    gguf_to_hf,
-)
-from executorch.backends.native.partitioner import NativePartitioner
-from executorch.backends.native.passes import get_default_passes
-from executorch.exir import to_edge_transform_and_lower
-from executorch.exir.lowered_backend_module import (
-    executorch_call_delegate,
-    get_lowered_submodules,
-)
+from executorch.backends.native.examples.hf.gguf_hf import gguf_architecture, gguf_to_hf
+from executorch.backends.native.examples.hf.hf_utils import report_lowering
 from executorch.exir.native import to_native
 from executorch.extension.llm.export.load import assign_one, iter_checkpoint, RawIter
 from torch.export import Dim, export
@@ -218,15 +206,6 @@ def _quantize_8da4w(
     )
 
 
-def _histogram(gm: torch.fx.GraphModule) -> collections.Counter:
-    return collections.Counter(
-        str(n.target)
-        for n in gm.graph.nodes
-        if n.op == "call_function"
-        and n.target not in (executorch_call_delegate, operator.getitem)
-    )
-
-
 def main() -> None:
     parser = argparse.ArgumentParser()
     parser.add_argument(
@@ -287,29 +266,7 @@ def main() -> None:
             dynamic_shapes={"input_ids": {1: seq}, "cache_position": {0: seq}},
             strict=not args.non_strict,
         )
-    print("== exported aten ops ==", flush=True)
-    for k, v in sorted(_histogram(ep.graph_module).items()):
-        print(f"  {v:4d} {k}")
-
-    edge = to_edge_transform_and_lower(
-        ep,
-        transform_passes=get_default_passes(),
-        partitioner=[
-            NativePartitioner(external_constants_tag=None, _serialize_as_ptn=True)
-        ],
-        compile_config=get_default_compile_config(),
-    )
-    gm = edge.exported_program().graph_module
-    lowered = get_lowered_submodules(gm)
-    print(f"== {len(lowered)} native delegate(s) ==")
-    delegated = collections.Counter()
-    for _, module, _ in lowered:
-        delegated += _histogram(module.original_module.graph_module)
-    for k, v in sorted(delegated.items()):
-        print(f"  {v:4d} {k}")
-    print("== residual (not delegated) ==")
-    for k, v in sorted(_histogram(gm).items()):
-        print(f"  {v:4d} {k}")
+    report_lowering(ep, "forward")
 
     if args.output:
         to_native(ep).save(args.output)

…/native/examples/hf_llm/export_hf_llm.py …ends/native/examples/hf/export_hf_llm.pybackends/native/examples/hf_llm/export_hf_llm.py renamed to backends/native/examples/hf/export_hf_llm.py backends/native/examples/hf_llm/export_hf_llm.py renamed to backends/native/examples/hf/export_hf_llm.py

@@ -0,0 +1,67 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the BSD-style license found in the
+# LICENSE file in the root directory of this source tree.
+
+"""Shared helpers for exporting Hugging Face models through the native backend.
+
+The per-model exporters in this directory (decoder-only LMs, Whisper) build
+their ``torch.export`` programs differently, but lower and report them the
+same way: an op histogram of what the native partitioner delegated versus
+what it left behind.
+"""
+
+import collections
+import operator
+
+import torch
+from executorch.backends.native import get_default_compile_config
+from executorch.backends.native.partitioner import NativePartitioner
+from executorch.backends.native.passes import get_default_passes
+from executorch.exir import to_edge_transform_and_lower
+from executorch.exir.lowered_backend_module import (
+    executorch_call_delegate,
+    get_lowered_submodules,
+)
+from torch.export import ExportedProgram
+
+
+def histogram(gm: torch.fx.GraphModule) -> collections.Counter:
+    """Count call_function targets, ignoring delegate plumbing."""
+    return collections.Counter(
+        str(n.target)
+        for n in gm.graph.nodes
+        if n.op == "call_function"
+        and n.target not in (executorch_call_delegate, operator.getitem)
+    )
+
+
+def report_lowering(ep: ExportedProgram, label: str) -> None:
+    """Lower ``ep`` for PTN and print delegated vs. residual op histograms."""
+    print(f"== {label}: exported aten ops ==", flush=True)
+    for k, v in sorted(histogram(ep.graph_module).items()):
+        print(f"  {v:4d} {k}")
+
+    edge = to_edge_transform_and_lower(
+        ep,
+        transform_passes=get_default_passes(),
+        partitioner=[
+            NativePartitioner(external_constants_tag=None, _serialize_as_ptn=True)
+        ],
+        compile_config=get_default_compile_config(),
+    )
+    gm = edge.exported_program().graph_module
+    lowered = get_lowered_submodules(gm)
+    print(f"== {label}: {len(lowered)} native delegate(s) ==")
+    delegated: collections.Counter = collections.Counter()
+    for _, module, _ in lowered:
+        delegated += histogram(module.original_module.graph_module)
+    for k, v in sorted(delegated.items()):
+        print(f"  {v:4d} {k}")
+    print(f"== {label}: residual (not delegated) ==")
+    residual = histogram(gm)
+    if not residual:
+        print("  (none)")
+    for k, v in sorted(residual.items()):
+        print(f"  {v:4d} {k}")

…ckends/native/examples/hf_llm/gguf_hf.py backends/native/examples/hf/gguf_hf.pybackends/native/examples/hf_llm/gguf_hf.py renamed to backends/native/examples/hf/gguf_hf.py backends/native/examples/hf_llm/gguf_hf.py renamed to backends/native/examples/hf/gguf_hf.py

@@ -18,8 +18,8 @@
 #include <utility>
 #include <vector>
 
-#include <executorch/backends/native/examples/hf_llm/runtime/LlmRunner.h>
-#include <executorch/backends/native/examples/hf_llm/runtime/LlmRunnerUtils.h>
+#include <executorch/backends/native/examples/hf/runtime/LlmRunner.h>
+#include <executorch/backends/native/examples/hf/runtime/LlmRunnerUtils.h>
 #include <executorch/backends/native/examples/runner_utils/runner_utils.h>
 #include <executorch/backends/native/runtime/Program.h>
 #include <executorch/backends/native/runtime/deserialize/Package.h>

backends/native/examples/hf/hf_utils.py

@@ -4,7 +4,7 @@
 // This source code is licensed under the BSD-style license found in the
 // LICENSE file in the root directory of this source tree.
 
-#include <executorch/backends/native/examples/hf_llm/runtime/LlmRunnerUtils.h>
+#include <executorch/backends/native/examples/hf/runtime/LlmRunnerUtils.h>
 
 #include <algorithm>
 #include <cctype>