Skip to content

Commit

Permalink
#13397: Add data parallel suppport for SqueezeBERT model
Browse files Browse the repository at this point in the history
  • Loading branch information
kkeerthana0573 committed Nov 21, 2024
1 parent 019a5cc commit 2f5a5b4
Show file tree
Hide file tree
Showing 9 changed files with 1,682 additions and 0 deletions.
33 changes: 33 additions & 0 deletions models/demos/wormhole/squeezebert/README.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,33 @@
# SqueezeBERT demo

Demo showcasing SqueezeBERT running on Grayskull - e150 and Wormhole - n150, n300 using ttnn.

## Introduction
SqueezeBERT is a bidirectional transformer similar to the BERT model. The key difference between the BERT architecture and the SqueezeBERT architecture is that SqueezeBERT uses grouped convolutions instead of fully-connected layers for the Q, K, V and FFN layers.


## Details
The entry point to functional_squeezebert model is squeezebert_for_question_answering in `models/demos/squeezebert/tt/ttnn_functional_squeezebert.py`. The model picks up certain configs and weights from huggingface pretrained model. We have used `squeezebert/squeezebert-uncased` version from huggingface as our reference.

### Sequence Size: 384
Sequence size determines the maximum length of input sequences processed by the model, optimizing performance and compatibility. It's recommended to set the sequence_size to 384

### Batch size: 16
Batch Size determines the number of input sequences processed simultaneously during training or inference, impacting computational efficiency and memory usage. On each device, the batch size will be 8, as the operations run in parallel. It's recommended to set the batch_size to 16

## How to Run

Use `pytest --disable-warnings models/demos/wormhole/squeezebert/demo/demo.py::test_demo[wormhole_b0-True-models.demos.wormhole.squeezebert.tt.ttnn_functional_squeezebert-squeezebert/squeezebert-uncased-models/demos/wormhole/squeezebert/demo/input_data.json-device_params0]` to run the demo.

If you wish to run the demo with a different input use `pytest --disable-warnings --input-path="<address_to_your_json_file.json>" models/demos/wormhole/squeezebert/demo/demo.py::test_demo[wormhole_b0-True-models.demos.wormhole.squeezebert.tt.ttnn_functional_squeezebert-squeezebert/squeezebert-uncased]`. This file is expected to have exactly 16 inputs.

Our second demo is designed to run SQuADV2 dataset, run this with `pytest --disable-warnings models/demos/wormhole/squeezebert/demo/demo.py::test_demo_squadv2[wormhole_b0-True-3-models.demos.wormhole.squeezebert.tt.ttnn_functional_squeezebert-squeezebert/squeezebert-uncased-device_params0]`.

If you wish to run for `n_iterations` samples, use `pytest --disable-warnings models/demos/wormhole/squeezebert/demo/demo.py::test_demo_squadv2[wormhole_b0-True-<n_iterations>-models.demos.wormhole.squeezebert.tt.ttnn_functional_squeezebert-squeezebert/squeezebert-uncased-device_params0]`


## Inputs
The demo receives inputs from respective `input_data.json` by default. To modify the inputs or specify a different path, adjust the input_path parameter in the command accordingly. It's recommended to avoid direct modifications to the input_data.json file.


#### Owner: [kkeerthana0573](https://github.com/kkeerthana0573)
314 changes: 314 additions & 0 deletions models/demos/wormhole/squeezebert/demo/demo.py
Original file line number Diff line number Diff line change
@@ -0,0 +1,314 @@
# SPDX-FileCopyrightText: © 2023 Tenstorrent Inc.

# SPDX-License-Identifier: Apache-2.0

import ttnn
import json
import torch
import pytest
import evaluate
import transformers
from loguru import logger
from models.utility_functions import (
profiler,
is_wormhole_b0,
skip_for_grayskull,
disable_compilation_reports,
disable_persistent_kernel_cache,
)
from ttnn.model_preprocessing import preprocess_model_parameters
from models.demos.wormhole.squeezebert.tt import ttnn_functional_squeezebert
from models.datasets.dataset_squadv2 import squadv2_1K_samples_input, squadv2_answer_decode_batch


def load_inputs(input_path, batch):
with open(input_path) as f:
input_data = json.load(f)
assert len(input_data) >= batch, f"Input data needs to have at least {batch} (batch size) entries."

context = []
question = []
for i in range(batch):
context.append(input_data[i]["context"])
question.append(input_data[i]["question"])

return context, question


def positional_ids(config, input_ids, past_key_values_length=0):
seq_length = input_ids.size(1)
position_ids = torch.arange(config.max_position_embeddings, dtype=torch.long, device=input_ids.device)
position_ids = position_ids.unsqueeze(0)[:, past_key_values_length : seq_length + past_key_values_length]
position_ids = position_ids.expand_as(input_ids)

return position_ids


def run_squeezebert_question_and_answering_inference(
mesh_device,
use_program_cache,
model_name,
batch_size,
sequence_size,
squeezebert,
input_path,
):
disable_persistent_kernel_cache()

hugging_face_reference_model = transformers.SqueezeBertForQuestionAnswering.from_pretrained(
model_name, torchscript=False
)

state_dict = hugging_face_reference_model.state_dict()
tt_model_name = f"ttnn_{model_name}_optimized"

profiler.start(f"preprocessing_parameter")
mesh_device_flag = is_wormhole_b0() and ttnn.GetNumAvailableDevices() == 2
batch_size = 16 if mesh_device_flag else 8
inputs_mesh_mapper = ttnn.ShardTensorToMesh(mesh_device, dim=0)
weights_mesh_mapper = ttnn.ReplicateTensorToMesh(mesh_device)
output_mesh_composer = ttnn.ConcatMeshToTensor(mesh_device, dim=0)
with ttnn.distribute(ttnn.ReplicateTensorToMesh(mesh_device)):
parameters = preprocess_model_parameters(
model_name=tt_model_name,
initialize_model=lambda: hugging_face_reference_model,
custom_preprocessor=ttnn_functional_squeezebert.custom_preprocessor,
device=mesh_device,
)
profiler.end(f"preprocessing_parameter")

tokenizer = transformers.SqueezeBertTokenizer.from_pretrained(model_name)
config = hugging_face_reference_model.config
nlp = transformers.pipeline("question-answering", model=hugging_face_reference_model, tokenizer=tokenizer)

context, question = load_inputs(input_path, batch_size)

preprocess_params, _, postprocess_params = nlp._sanitize_parameters()
preprocess_params["max_seq_len"] = sequence_size
inputs = nlp._args_parser({"context": context, "question": question})

preprocessed_inputs = []
for i in range(batch_size):
model_input = next(nlp.preprocess(inputs[0][i], **preprocess_params))

single_input = {
"example": model_input["example"],
"inputs": model_input,
}
preprocessed_inputs.append(single_input)

squeezebert_input = tokenizer.batch_encode_plus(
zip(question, context),
max_length=sequence_size,
padding="max_length",
truncation=True,
return_attention_mask=True,
return_token_type_ids=True,
return_tensors="pt",
)

profiler.start(f"preprocessing_input")
position_ids = positional_ids(config, squeezebert_input.input_ids)
ttnn_squeezebert_inputs = squeezebert.preprocess_inputs(
squeezebert_input["input_ids"],
squeezebert_input["token_type_ids"],
position_ids,
squeezebert_input["attention_mask"],
device=mesh_device,
mesh_mapper=inputs_mesh_mapper,
)
profiler.end(f"preprocessing_input")

profiler.start(f"inference_time")
tt_output = squeezebert.squeezebert_for_question_answering(
config,
*ttnn_squeezebert_inputs,
state_dict=state_dict,
base_addr=f"transformer.",
parameters=parameters,
device=mesh_device,
reader_patterns_cache={},
mesh_mapper=inputs_mesh_mapper,
mesh_composer=output_mesh_composer,
)
profiler.end(f"inference_time")

tt_output = ttnn.to_torch(tt_output, mesh_composer=output_mesh_composer).to(torch.float32)

tt_start_logits = tt_output[..., :, 0].squeeze(1)
tt_end_logits = tt_output[..., :, 1].squeeze(1)

model_answers = {}
profiler.start("post_processing_output_to_string")
for i in range(batch_size):
tt_res = {
"start": tt_start_logits[i],
"end": tt_end_logits[i],
"example": preprocessed_inputs[i]["example"],
**preprocessed_inputs[i]["inputs"],
}
tt_answer = nlp.postprocess([tt_res], **postprocess_params)

logger.info(f"answer: {tt_answer['answer']}\n")
model_answers[i] = tt_answer["answer"]

profiler.end("post_processing_output_to_string")

measurements = {
"preprocessing_parameter": profiler.get("preprocessing_parameter"),
"preprocessing_input": profiler.get("preprocessing_input"),
"inference_time": profiler.get("inference_time"),
"post_processing": profiler.get("post_processing_output_to_string"),
}
logger.info(f"preprocessing_parameter: {measurements['preprocessing_parameter']} s")
logger.info(f"preprocessing_input: {measurements['preprocessing_input']} s")
logger.info(f"inference_time: {measurements['inference_time']} s")
logger.info(f"post_processing : {measurements['post_processing']} s")

return measurements


def run_squeezebert_question_and_answering_inference_squad_v2(
mesh_device,
use_program_cache,
model_name,
batch_size,
sequence_size,
squeezebert,
n_iterations,
):
disable_persistent_kernel_cache()
hugging_face_reference_model = transformers.SqueezeBertForQuestionAnswering.from_pretrained(
model_name, torchscript=False
)

state_dict = hugging_face_reference_model.state_dict()
tt_model_name = f"ttnn_{model_name}_optimized"

mesh_device_flag = is_wormhole_b0() and ttnn.GetNumAvailableDevices() == 2
batch_size = 16 if mesh_device_flag else 8
inputs_mesh_mapper = ttnn.ShardTensorToMesh(mesh_device, dim=0)
weights_mesh_mapper = ttnn.ReplicateTensorToMesh(mesh_device)
output_mesh_composer = ttnn.ConcatMeshToTensor(mesh_device, dim=0)
with ttnn.distribute(ttnn.ReplicateTensorToMesh(mesh_device)):
parameters = preprocess_model_parameters(
model_name=tt_model_name,
initialize_model=lambda: hugging_face_reference_model,
custom_preprocessor=ttnn_functional_squeezebert.custom_preprocessor,
device=mesh_device,
)

tokenizer = transformers.SqueezeBertTokenizer.from_pretrained(model_name)
config = hugging_face_reference_model.config

nlp = transformers.pipeline("question-answering", model=hugging_face_reference_model, tokenizer=tokenizer)

attention_mask = True
token_type_ids = True
inputs_squadv2 = squadv2_1K_samples_input(tokenizer, sequence_size, attention_mask, token_type_ids, batch_size)
squad_metric = evaluate.load("squad_v2")

with torch.no_grad():
pred_labels = []
cpu_pred_labels = []
true_labels = []
i = 0
for batch in inputs_squadv2:
if i < n_iterations:
batch_data = batch[0]
curr_batch_size = batch_data["input_ids"].shape[0]
position_ids = positional_ids(config, batch_data.input_ids)

ttnn_squeezebert_inputs = squeezebert.preprocess_inputs(
batch_data["input_ids"],
batch_data["token_type_ids"],
position_ids,
batch_data["attention_mask"],
device=mesh_device,
mesh_mapper=inputs_mesh_mapper,
)

tt_output = squeezebert.squeezebert_for_question_answering(
config,
*ttnn_squeezebert_inputs,
state_dict=state_dict,
base_addr=f"transformer.",
parameters=parameters,
device=mesh_device,
reader_patterns_cache={},
mesh_mapper=inputs_mesh_mapper,
mesh_composer=output_mesh_composer,
)
tt_output = ttnn.to_torch(tt_output, mesh_composer=output_mesh_composer).to(torch.float32)

cpu_output = hugging_face_reference_model(**batch_data)
references = batch[1]
question = batch[2]
context = batch[3]

cpu_predictions, tt_predictions = squadv2_answer_decode_batch(
hugging_face_reference_model,
tokenizer,
nlp,
references,
cpu_output,
tt_output,
curr_batch_size,
question,
context,
)
pred_labels.extend(tt_predictions)
cpu_pred_labels.extend(cpu_predictions)
true_labels.extend(references)

del tt_output
i += 1
eval_score = squad_metric.compute(predictions=pred_labels, references=true_labels)
cpu_eval_score = squad_metric.compute(predictions=cpu_pred_labels, references=true_labels)
logger.info(f"\tTT_Eval: exact: {eval_score['exact']} -- F1: {eval_score['f1']}")
# logger.info(f"\tCPU_Eval: exact: {cpu_eval_score['exact']} -- F1: {cpu_eval_score['f1']}")


@skip_for_grayskull()
@pytest.mark.parametrize("device_params", [{"l1_small_size": 16384}], indirect=True)
@pytest.mark.parametrize(
"model_name, input_loc",
((["squeezebert/squeezebert-uncased", "models/demos/wormhole/squeezebert/demo/input_data.json"]),),
)
@pytest.mark.parametrize("squeezebert", [ttnn_functional_squeezebert])
def test_demo(input_loc, model_name, squeezebert, mesh_device, use_program_cache, reset_seeds):
disable_persistent_kernel_cache()
disable_compilation_reports()

return run_squeezebert_question_and_answering_inference(
mesh_device=mesh_device,
use_program_cache=use_program_cache,
model_name=model_name,
batch_size=8,
sequence_size=384,
squeezebert=squeezebert,
input_path=input_loc,
)


@skip_for_grayskull()
@pytest.mark.parametrize("device_params", [{"l1_small_size": 16384}], indirect=True)
@pytest.mark.parametrize("model_name", ["squeezebert/squeezebert-uncased"])
@pytest.mark.parametrize("squeezebert", [ttnn_functional_squeezebert])
@pytest.mark.parametrize(
"n_iterations",
((3),),
)
def test_demo_squadv2(model_name, squeezebert, n_iterations, mesh_device, use_program_cache, reset_seeds):
disable_persistent_kernel_cache()
disable_compilation_reports()

return run_squeezebert_question_and_answering_inference_squad_v2(
mesh_device=mesh_device,
use_program_cache=use_program_cache,
model_name=model_name,
batch_size=8,
sequence_size=384,
squeezebert=squeezebert,
n_iterations=n_iterations,
)
Loading

0 comments on commit 2f5a5b4

Please sign in to comment.