Add support for save quantized checkpoint in llama code (#553)

Summary:
The goal is to upload a torchao quantized model to huggingface so that we can run the model in huggingface

Test Plan:
python generate.py -q int4wo-32 --save

Reviewers:

Subscribers:

Tasks:

Tags:

scripts/hf_eval.py

@@ -40,12 +40,12 @@ def format_value(value):
 
     print(tabulate(main_table, headers=['Task', 'Metrics'], tablefmt='grid'))
 
-def run_evaluation(repo_id, tasks, limit, device, precision, quantization, compile, batch_size, max_length):
+def run_evaluation(repo_id, tasks, limit, device, precision, quantization, compile, save, batch_size, max_length):
 
     tokenizer = AutoTokenizer.from_pretrained(repo_id)
     model = AutoModelForCausalLM.from_pretrained(repo_id).to(device="cpu", dtype=precision)
 
-    if compile:
+    if quantization == "autoquant" and compile:
         model = torch.compile(model, mode="max-autotune", fullgraph=True)
 
     if quantization == "int8dq":
@@ -57,6 +57,10 @@ def run_evaluation(repo_id, tasks, limit, device, precision, quantization, compi
         quantize_(model.to(device=device), int4_weight_only())
     elif quantization == "autoquant":
         model = autoquant(model.to(device=device))
+
+    if quantization != "autoquant" and compile:
+        model = torch.compile(model, mode="max-autotune", fullgraph=True)
+
     with torch.no_grad():
         result = evaluate(
             HFLM(
@@ -70,6 +74,12 @@ def run_evaluation(repo_id, tasks, limit, device, precision, quantization, compi
 
         pretty_print_nested_results(result)
 
+    if save:
+        # This doesn't work yet: https://github.com/huggingface/transformers/issues/32364
+        # model.save_pretrained("quantized_model_test", safe_serialization=False)
+        file_name = repo_id.split("/")[-1] + "-" + quantization + ".pt"
+        torch.save(model.state_dict(), file_name)
+
 
 if __name__ == '__main__':
     import argparse
@@ -81,8 +91,9 @@ def run_evaluation(repo_id, tasks, limit, device, precision, quantization, compi
     parser.add_argument('--device', type=str, default="cuda", help='Device to use for evaluation')
     parser.add_argument('-q', '--quantization', default = "None", choices=["int8dq", "int8wo", "int4wo","autoquant", "None"], help='Which quantization technique to apply')
     parser.add_argument('--compile', action='store_true', help='Whether to compile the model.')
+    parser.add_argument('--save', action='store_true', help='Whether to save the model.')
     parser.add_argument('--batch_size', type=int, default=1, help='Batch size to use for evaluation, note int8wo and int4wo work best with small batchsizes, int8dq works better with large batchsizes')
     parser.add_argument('--max_length', type=int, default=None, help='Length of text to process at one time')
 
     args = parser.parse_args()
-    run_evaluation(args.repo_id, args.tasks, args.limit, args.device, args.precision, args.quantization, args.compile, args.batch_size, args.max_length)
+    run_evaluation(args.repo_id, args.tasks, args.limit, args.device, args.precision, args.quantization, args.compile, args.save, args.batch_size, args.max_length)

torchao/_models/llama/generate.py

@@ -3,6 +3,7 @@
 
 # This source code is licensed under the license found in the
 # LICENSE file in the root directory of this source tree.
+import os
 import sys
 import time
 from pathlib import Path
@@ -165,6 +166,7 @@ def main(
     checkpoint_path: Path = Path("checkpoints/meta-Transformer/Transformer-2-7b-chat-hf/model.pth"),
     quantization: Optional[str] = None,
     kv_cache_quantization: bool = False,
+    save: bool = False,
     compile: bool = True,
     compile_prefill: bool = False,
     profile: Optional[Path] = None,
@@ -238,6 +240,11 @@ def main(
 
     model_size = get_model_size_in_bytes(model, ignore_embeddings=True) / 1e9
 
+    if save:
+        output_dir = str(checkpoint_path.cwd())
+        filename = str(checkpoint_path.name).split(".")[0]
+        torch.save(model.state_dict(), os.path.join(output_dir, filename + f"-{quantization}.pt"))
+
     if compile:
         print("Compiling Model")
         global decode_one_token, prefill
@@ -362,6 +369,7 @@ def callback(x):
     parser.add_argument('--checkpoint_path', type=Path, default=Path("../../../checkpoints/meta-llama/Llama-2-7b-chat-hf/model.pth"), help='Model checkpoint path.')
     parser.add_argument('-q', '--quantization', type=str, help='Which quantization techniques to apply: int8dq, int8wo, int4wo-<groupsize>, autoquant')
     parser.add_argument('--kv_cache_quantization', action='store_true', help='Whether to quantize the KV cache')
+    parser.add_argument('--save', action='store_true', help='Whether to save the quantized model.')
     parser.add_argument('--compile', action='store_true', help='Whether to compile the model.')
     parser.add_argument('--compile_prefill', action='store_true', help='Whether to compile the prefill (improves prefill perf, but higher compile times)')
     parser.add_argument('--profile', type=Path, default=None, help='Profile path.')
@@ -372,5 +380,5 @@ def callback(x):
     args = parser.parse_args()
     main(
         args.prompt, args.interactive, args.num_samples, args.max_new_tokens, args.top_k,
-        args.temperature, args.checkpoint_path, args.quantization, args.kv_cache_quantization, args.compile, args.compile_prefill, args.profile, args.device, args.precision, args.write_result
+        args.temperature, args.checkpoint_path, args.quantization, args.kv_cache_quantization, args.save, args.compile, args.compile_prefill, args.profile, args.device, args.precision, args.write_result
     )