Add Intel AutoRound algorithm support (#1994)

Resolve #1968


### Highlights
- Introduced `AutoRoundModifier` to enable AutoRound quantization for
`wNa16`.
- Added an end-to-end example and unit tests.
- Verified functionality with local accuracy tests (GSM8K with a limit
of 1000, the results may fluctuate due to non-determinism.)
```bash
- LLMC-AutoRound
vllm (pretrained=/storage/yiliu7/Meta-Llama-3-8B-Instruct-W4A16-G128-disbale-shuffule,tensor_parallel_size=1,max_model_len=8192,max_num_batched_tokens=32768,max_num_seqs=128,add_bos_token=True,gpu_memory_utilization=0.8,dtype=bfloat16,max_gen_toks=2048,enable_prefix_caching=False), gen_kwargs: (None), limit: 1000.0, num_fewshot: None, batch_size: 128
|Tasks|Version|     Filter     |n-shot|  Metric   |   |Value|   |Stderr|
|-----|------:|----------------|-----:|-----------|---|----:|---|-----:|
|gsm8k|      3|flexible-extract|     5|exact_match|↑  |0.737|±  |0.0139|
|     |       |strict-match    |     5|exact_match|↑  |0.736|±  |0.0139|

- AutoRound result as ref
vllm (pretrained=/storage/yiliu7/meta-llama/Meta-Llama-3-8B-Instruct-ar/Meta-Llama-3-8B-Instruct-w4g128/,tensor_parallel_size=1,max_model_len=8192,max_num_batched_tokens=32768,max_num_seqs=128,add_bos_token=True,gpu_memory_utilization=0.8,dtype=bfloat16,max_gen_toks=2048,enable_prefix_caching=False), gen_kwargs: (None), limit: 1000.0, num_fewshot: None, batch_size: 128
|Tasks|Version|     Filter     |n-shot|  Metric   |   |Value|   |Stderr|
|-----|------:|----------------|-----:|-----------|---|----:|---|-----:|
|gsm8k|      3|flexible-extract|     5|exact_match|↑  |0.739|±  |0.0139|
|     |       |strict-match    |     5|exact_match|↑  |0.740|±  |0.0139|

```
Attached [eval
cmd](https://gist.github.com/yiliu30/a7881cd1cbf0d676e3ffac3e3833aa8e)
FYI.


### Next stage (in later PRs)
- [ ] Extend support for additional data types.
- [ ] Add group-wise quantization recipes mapping between LLMC and
AutoRound.
- [ ] Add end-to-end tests.

cc @hshen14 @thuang6 @wenhuach21

---------

Signed-off-by: yiliu30 <yi4.liu@intel.com>
Signed-off-by: Yi Liu <yi4.liu@intel.com>
Co-authored-by: Brian Dellabetta <brian-dellabetta@users.noreply.github.com>

Author: yiliu30

.github/workflows/test-check-transformers.yaml

@@ -97,6 +97,10 @@ jobs:
         if: (success() || failure()) && steps.install.outcome == 'success'
         run: |
           pytest -v tests/llmcompressor/transformers/gptq
+      - name: Running AutoRound Tests
+        if: (success() || failure()) && steps.install.outcome == 'success'
+        run: |
+          pytest -v tests/llmcompressor/transformers/autoround
       - name: Running ONESHOT Tests
         if: (success() || failure()) && steps.install.outcome == 'success'
         run: |

examples/autoround/llama3_example.py

@@ -0,0 +1,56 @@
+from auto_round.calib_dataset import get_dataset
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+from llmcompressor import oneshot
+from llmcompressor.modifiers.autoround import AutoRoundModifier
+from llmcompressor.utils import dispatch_for_generation
+
+# Select model and load it.
+model_id = "meta-llama/Meta-Llama-3-8B-Instruct"
+model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype="auto")
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+
+# Select calibration dataset.
+NUM_CALIBRATION_SAMPLES = 128
+MAX_SEQUENCE_LENGTH = 2048
+# Get aligned calibration dataset.
+
+ds = get_dataset(
+    tokenizer=tokenizer,
+    seqlen=MAX_SEQUENCE_LENGTH,
+    nsamples=NUM_CALIBRATION_SAMPLES,
+)
+
+
+# Configure the quantization algorithm to run.
+#   * quantize the weights to 4 bit with AutoRound with a group size 128
+recipe = AutoRoundModifier(
+    targets="Linear", scheme="W4A16", ignore=["lm_head"], iters=200
+)
+
+
+# Apply algorithms.
+oneshot(
+    model=model,
+    dataset=ds,
+    recipe=recipe,
+    max_seq_length=MAX_SEQUENCE_LENGTH,
+    num_calibration_samples=NUM_CALIBRATION_SAMPLES,
+    # disable shuffling to get slightly better mmlu score
+    shuffle_calibration_samples=False,
+)
+
+# Confirm generations of the quantized model look sane.
+print("\n\n")
+print("========== SAMPLE GENERATION ==============")
+dispatch_for_generation(model)
+sample = tokenizer("Hello my name is", return_tensors="pt")
+sample = {key: value.to(model.device) for key, value in sample.items()}
+output = model.generate(**sample, max_new_tokens=100)
+print(tokenizer.decode(output[0]))
+print("==========================================\n\n")
+
+# Save to disk compressed.
+SAVE_DIR = model_id.rstrip("/").split("/")[-1] + "-W4A16-G128-AutoRound"
+model.save_pretrained(SAVE_DIR, save_compressed=True)
+tokenizer.save_pretrained(SAVE_DIR)

setup.py

@@ -144,6 +144,8 @@ def localversion_func(version: ScmVersion) -> str:
             if BUILD_TYPE == "release"
             else "compressed-tensors>=0.12.3a2"
         ),
+        # TODO: replace it with the release version
+        ("auto_round @ git+https://github.com/intel/auto-round.git@llmc"),
     ],
     extras_require={
         "dev": [

src/llmcompressor/modifiers/autoround/__init__.py

@@ -0,0 +1,3 @@
+# ruff: noqa
+
+from .base import *