def test_nested3(self):
r"""More complicated nested test case with child qconfig overrides
parent qconfig
"""
model = AnnotatedCustomConfigNestedModel()
prepare(model)
def checkPrepModules(model, before_calib=False):
if before_calib:
self.checkObservers(model)
self.checkNoPrepModules(model)
self.checkNoPrepModules(model.sub1)
self.checkNoPrepModules(model.sub1.fc)
self.checkNoPrepModules(model.sub1.relu)
self.checkNoPrepModules(model.sub2)
self.checkHasPrepModules(model.sub2.fc1)
self.checkHasPrepModules(model.sub2.fc2)
self.checkHasPrepModules(model.fc3)
checkPrepModules(model, True)
test_only_eval_fn(model, self.calib_data)
convert(model)
def checkQuantized(model):
checkPrepModules(model)
self.checkWrappedQuantizedLinear(model.sub2.fc1)
self.checkWrappedQuantizedLinear(model.sub2.fc2)
self.checkWrappedQuantizedLinear(model.fc3)
test_only_eval_fn(model, self.calib_data)
self.checkScriptable(model, self.calib_data)
checkQuantized(model)
# test one line API
model = quantize(AnnotatedCustomConfigNestedModel(), test_only_eval_fn,
self.calib_data)
checkQuantized(model)
def test_nested1(self):
r"""Test quantization for nested model, top level 'fc3' and
'fc1' of submodule 'sub2', 'sub2.fc2' is not quantized
"""
model = NestedModel().eval()
qconfig_dict = {'fc3': default_qconfig, 'sub2.fc1': default_qconfig}
def checkPrepModules(model, before_calib=False):
if before_calib:
self.checkObservers(model)
self.checkNoPrepModules(model)
self.checkNoPrepModules(model.sub1)
self.checkNoPrepModules(model.sub1.fc)
self.checkNoPrepModules(model.sub1.relu)
self.checkNoPrepModules(model.sub2)
self.checkHasPrepModules(model.sub2.fc1)
self.checkNoPrepModules(model.sub2.fc2)
self.checkHasPrepModules(model.fc3)
model = prepare(model, qconfig_dict)
checkPrepModules(model, True)
test_only_eval_fn(model, self.calib_data)
convert(model)
def checkQuantized(model):
checkPrepModules(model)
self.checkLinear(model.sub1.fc)
self.checkQuantizedLinear(model.fc3)
self.checkQuantizedLinear(model.sub2.fc1)
self.checkLinear(model.sub2.fc2)
test_only_eval_fn(model, self.calib_data)
checkQuantized(model)
# test one line API
model = quantize(NestedModel().eval(), test_only_eval_fn,
self.calib_data, qconfig_dict)
checkQuantized(model)
def test_nested1(self):
r"""Test quantization for nested model, top level 'fc3' and
'fc1' of submodule 'sub2', 'sub2.fc2' is not quantized
"""
model = AnnotatedNestedModel()
def checkPrepModules(model, before_calib=False):
if before_calib:
self.checkObservers(model)
self.checkNoPrepModules(model)
self.checkNoPrepModules(model.sub1)
self.checkNoPrepModules(model.sub1.fc)
self.checkNoPrepModules(model.sub1.relu)
self.checkNoPrepModules(model.sub2)
self.checkHasPrepModules(model.sub2.fc1)
self.checkNoPrepModules(model.sub2.fc2)
self.checkHasPrepModules(model.fc3)
prepare(model)
checkPrepModules(model, True)
test_only_eval_fn(model, self.calib_data)
convert(model)
def checkQuantized(model):
checkPrepModules(model)
self.checkLinear(model.sub1.fc)
self.checkWrappedQuantizedLinear(model.fc3)
self.checkWrappedQuantizedLinear(model.sub2.fc1)
self.checkLinear(model.sub2.fc2)
test_only_eval_fn(model, self.calib_data)
self.checkScriptable(model, self.calib_data)
checkQuantized(model)
# test one line API
model = quantize(AnnotatedNestedModel(), test_only_eval_fn,
self.calib_data)
checkQuantized(model)
def test_manual(self):
r"""User inserts QuantStub and DeQuantStub in model code
and call the quantization utility functions.
"""
model = QuantStubModel()
# propagate the qconfig of parents to children, model is changed
# inplace
prepare(model)
self.checkObservers(model)
test_only_eval_fn(model, self.calib_data)
convert(model)
def checkQuantized(model):
self.assertEqual(type(model.fc), nnq.Linear)
test_only_eval_fn(model, self.calib_data)
self.checkScriptable(model, self.calib_data)
checkQuantized(model)
# test one line API
model = quantize(QuantStubModel(), test_only_eval_fn, self.calib_data)
checkQuantized(model)
def test_nested2(self):
model = AnnotatedSubNestedModel()
prepare(model)
def checkPrepModules(model, before_calib=False):
if before_calib:
self.checkObservers(model)
self.checkNoPrepModules(model)
self.checkNoPrepModules(model.sub1)
self.checkNoPrepModules(model.sub1.fc)
self.checkNoPrepModules(model.sub1.relu)
self.checkHasPrepModules(model.sub2)
self.checkNoPrepModules(model.sub2.module.fc1)
self.checkNoPrepModules(model.sub2.module.fc2)
self.checkHasPrepModules(model.fc3)
checkPrepModules(model, True)
test_only_eval_fn(model, self.calib_data)
convert(model)
def checkQuantized(model):
checkPrepModules(model)
self.checkLinear(model.sub1.fc)
self.assertEqual(type(model.sub1.relu), torch.nn.ReLU)
self.checkQuantizedLinear(model.sub2.module.fc1)
self.checkQuantizedLinear(model.sub2.module.fc2)
self.checkWrappedQuantizedLinear(model.fc3)
test_only_eval_fn(model, self.calib_data)
self.checkScriptable(model, self.calib_data)
checkQuantized(model)
# test one line API
model = quantize(AnnotatedSubNestedModel(), test_only_eval_fn,
self.calib_data)
checkQuantized(model)
def test_skip_quant(self):
r"""The case when we want to skip quantizing some layers
"""
model = SkipQuantModel()
prepare(model)
self.checkObservers(model)
test_only_eval_fn(model, self.calib_data)
convert(model)
def checkQuantized(model):
self.checkLinear(model.fc)
self.checkQuantDequant(model.sub)
self.checkQuantizedLinear(model.sub.module.fc1)
self.checkQuantizedLinear(model.sub.module.fc2)
self.assertEqual(type(model.sub.module.relu), nnq.ReLU)
self.checkScriptable(model, self.calib_data)
checkQuantized(model)
# test one line API
model = quantize(SkipQuantModel(), test_only_eval_fn, self.calib_data)
checkQuantized(model)
def test_fuse_module_eval(self):
model = ModelForFusion(default_qconfig)
model.eval()
fuse_modules(model,
[['conv1', 'bn1', 'relu1'], ['sub1.conv', 'sub1.bn']])
self.assertEqual(type(model.conv1), nni.ConvReLU2d,
"Fused Conv + BN + Relu first layer (BN is folded)")
self.assertEqual(type(model.conv1[0]), nn.Conv2d,
"Fused Conv + BN + Relu (Conv + folded BN only)")
self.assertEqual(type(model.conv1[1]), nn.ReLU,
"Fused Conv + BN + Relu second layer (Relu only)")
self.assertEqual(type(model.bn1), nn.Identity,
"Fused Conv + BN + Relu second layer (Skipped BN)")
self.assertEqual(type(model.relu1), nn.Identity,
"Fused Conv + BN + Relu second layer (Skipped Relu)")
self.assertEqual(type(model.sub1.conv), nn.Conv2d,
"Fused submodule Conv + folded BN")
self.assertEqual(type(model.sub1.bn), nn.Identity,
"Fused submodule (skipped BN)")
self.assertEqual(type(model.sub2.conv), nn.Conv2d,
"Non-fused submodule Conv")
self.assertEqual(type(model.sub2.relu), torch.nn.ReLU,
"Non-fused submodule ReLU")
model = prepare(model)
self.checkObservers(model)
test_only_eval_fn(model, self.img_data)
model = convert(model)
def checkQuantized(model):
self.assertEqual(type(model.conv1), nniq.ConvReLU2d)
self.assertEqual(type(model.bn1), nn.Identity)
self.assertEqual(type(model.relu1), nn.Identity)
self.assertEqual(type(model.sub1.conv), nnq.Conv2d)
self.assertEqual(type(model.sub1.bn), nn.Identity)
self.assertEqual(type(model.sub2.conv), nn.Conv2d)
self.assertEqual(type(model.sub2.relu), nn.ReLU)
test_only_eval_fn(model, self.img_data)
checkQuantized(model)
model = ModelForFusion(default_qconfig).eval()
fuse_modules(model,
[['conv1', 'bn1', 'relu1'], ['sub1.conv', 'sub1.bn']])
model = quantize(model, test_only_eval_fn, self.img_data)
checkQuantized(model)
def quantize(self):
"""Quantize the model and save as self.quantizedmodel.
Weights are quantized to 8-bit precision and the model is saved as
quantizemodel.pth in self.model_dir directory.
Returns
-------
[type]
[description]
"""
print("Quantizing model..")
self.learn.model.cpu()
self.quantizedmodel = convert(self.learn.model, inplace=False)
torch.save(self.quantizedmodel, self.model_dir / "quantizedmodel.pth")
self.learn.model.to("cuda")
return self.quantizedmodel
def test_relu(self):
class M(torch.nn.Module):
def __init__(self):
super().__init__()
self.relu = nn.ReLU()
def forward(self, x):
x = self.relu(x)
return x
m = M().train()
m.qconfig = default_qconfig
m = prepare_qat(m)
# make sure no activation_post_process is inserted for relu
self.assertFalse(hasattr(m, "activation_post_process"))
m = convert(m)
# make sure ReLU module is not changed
self.assertTrue(type(m.relu), nn.ReLU)
def test_manual(self):
model = ManualLinearQATModel()
model = prepare_qat(model)
self.checkObservers(model)
test_only_train_fn(model, self.train_data)
model = convert(model)
def checkQuantized(model):
self.assertEqual(type(model.fc1), nnq.Linear)
self.assertEqual(type(model.fc2), nnq.Linear)
test_only_eval_fn(model, self.calib_data)
self.checkScriptable(model, self.calib_data)
checkQuantized(model)
model = quantize_qat(ManualLinearQATModel(), test_only_train_fn,
self.train_data)
checkQuantized(model)
def load(checkpoint_dir, model):
"""Execute the quantize process on the specified model.
Args:
checkpoint_dir (dir): The folder of checkpoint.
'best_configure.yaml' and 'best_model_weights.pt' are needed
in This directory. 'checkpoint' dir is under workspace folder
and workspace folder is define in configure yaml file.
model (object): fp32 model need to do quantization.
Returns:
(object): quantized model
"""
tune_cfg_file = os.path.join(
os.path.abspath(os.path.expanduser(checkpoint_dir)),
'best_configure.yaml')
weights_file = os.path.join(
os.path.abspath(os.path.expanduser(checkpoint_dir)),
'best_model_weights.pt')
assert os.path.exists(
tune_cfg_file), "tune configure file %s didn't exist" % tune_cfg_file
assert os.path.exists(
weights_file), "weight file %s didn't exist" % weights_file
q_model = copy.deepcopy(model.eval())
with open(tune_cfg_file, 'r') as f:
tune_cfg = yaml.load(f, Loader=yaml.UnsafeLoader)
op_cfgs = _cfg_to_qconfig(tune_cfg)
_propagate_qconfig(q_model, op_cfgs)
# sanity check common API misusage
if not any(hasattr(m, 'qconfig') and m.qconfig for m in q_model.modules()):
logger.warn(
"None of the submodule got qconfig applied. Make sure you "
"passed correct configuration through `qconfig_dict` or "
"by assigning the `.qconfig` attribute directly on submodules")
add_observer_(q_model)
q_model = convert(q_model, inplace=True)
weights = torch.load(weights_file)
q_model.load_state_dict(weights)
return q_model
def test_resnet_base(self, qconfig):
r"""Test quantization for bottleneck topology used in resnet/resnext
and add coverage for conversion of average pool and float functional
"""
model = ResNetBase().float().eval()
model = QuantWrapper(model)
model.qconfig = qconfig
fuse_list = ['module.conv1', 'module.bn1', 'module.relu1']
fuse_modules(model, fuse_list, inplace=True)
model = prepare(model)
self.checkObservers(model)
test_only_eval_fn(model, self.img_data)
model = convert(model)
def checkQuantized(model):
self.assertEqual(type(model.module.conv1), nn._intrinsic.quantized.ConvReLU2d)
self.assertEqual(type(model.module.myop), nn.quantized.QFunctional)
self.assertEqual(type(model.module.avgpool), nn.AdaptiveAvgPool2d)
test_only_eval_fn(model, self.img_data)
checkQuantized(model)
def _test_activation_convert_numerics_impl(self, Act, data):
class M(torch.nn.Module):
def __init__(self):
super().__init__()
self.act = Act()
self.quant = QuantStub()
self.dequant = DeQuantStub()
def forward(self, x):
x = self.quant(x)
x = self.act(x)
x = self.dequant(x)
return x
m = M().train()
m.qconfig = default_qat_qconfig
m = prepare_qat(m)
before_convert = m(data)
m = convert(m)
after_convert = m(data)
self.assertEqual(before_convert, after_convert)
def test_manual(self):
for qengine in supported_qengines:
with override_quantized_engine(qengine):
model = ManualLinearQATModel(qengine)
model = prepare_qat(model)
self.checkObservers(model)
test_only_train_fn(model, self.train_data)
model = convert(model)
def checkQuantized(model):
self.assertEqual(type(model.fc1), nnq.Linear)
self.assertEqual(type(model.fc2), nnq.Linear)
test_only_eval_fn(model, self.calib_data)
self.checkScriptable(model, self.calib_data)
self.checkNoQconfig(model)
checkQuantized(model)
model = quantize_qat(ManualLinearQATModel(qengine),
test_only_train_fn, [self.train_data])
checkQuantized(model)
def test_conv_linear(self):
model = ManualConvLinearQATModel()
model = prepare_qat(model)
self.checkObservers(model)
test_only_train_fn(model, self.img_data)
model = convert(model)
def checkQuantized(model):
self.assertEqual(type(model.conv), nnq.Conv2d)
self.assertEqual(type(model.fc1), nnq.Linear)
self.assertEqual(type(model.fc2), nnq.Linear)
test_only_eval_fn(model, self.img_data)
self.checkScriptable(model, self.img_data)
checkQuantized(model)
model = ManualConvLinearQATModel()
model = quantize_qat(model, test_only_train_fn, self.img_data)
checkQuantized(model)
def test_single_layer(self, qconfig):
r"""Quantize SingleLayerLinearModel which has one Linear module, make sure it is swapped
to nnq.Linear which is the quantized version of the module
"""
model = AnnotatedSingleLayerLinearModel()
model.qconfig = qconfig
model = prepare(model)
# Check if observers and quant/dequant nodes are inserted
self.checkNoPrepModules(model)
self.checkHasPrepModules(model.fc1)
self.checkObservers(model)
test_only_eval_fn(model, self.calib_data)
model = convert(model)
def checkQuantized(model):
self.checkNoPrepModules(model)
self.checkHasPrepModules(model.fc1)
self.checkWrappedQuantizedLinear(model.fc1)
test_only_eval_fn(model, self.calib_data)
self.checkScriptable(model, self.calib_data)
checkQuantized(model)
# test one line API - out of place version
base = AnnotatedSingleLayerLinearModel()
base.qconfig = qconfig
keys_before = set(list(base.state_dict().keys()))
model = quantize(base, test_only_eval_fn, self.calib_data)
checkQuantized(model)
keys_after = set(list(base.state_dict().keys()))
self.assertEqual(keys_before,
keys_after) # simple check that nothing changed
# in-place version
model = AnnotatedSingleLayerLinearModel()
model.qconfig = qconfig
quantize(model, test_only_eval_fn, self.calib_data, inplace=True)
checkQuantized(model)
def test_compare_model_stub_functional_static(self):
r"""Compare the output of static quantized functional layer and its float shadow module
"""
qengine = torch.backends.quantized.engine
model = ModelWithFunctionals().eval()
model.qconfig = torch.quantization.get_default_qconfig("fbgemm")
q_model = prepare(model, inplace=False)
q_model(self.img_data_2d[0][0])
q_model = convert(q_model)
module_swap_list = [nnq.FloatFunctional]
ob_dict = compare_model_stub(model, q_model, module_swap_list,
self.img_data_2d[0][0])
self.assertEqual(len(ob_dict), 6)
self.assertTrue(isinstance(q_model.mycat, Shadow))
self.assertTrue(isinstance(q_model.myadd, Shadow))
self.assertTrue(isinstance(q_model.mymul, Shadow))
self.assertTrue(isinstance(q_model.myadd_relu, Shadow))
self.assertTrue(isinstance(q_model.my_scalar_add, Shadow))
self.assertTrue(isinstance(q_model.my_scalar_mul, Shadow))
for k, v in ob_dict.items():
self.assertTrue(v["float"].shape == v["quantized"].shape)
def train_subject_specific_quant(subject,
epochs=500,
batch_size=32,
lr=0.001,
silent=False,
plot=True,
**kwargs):
"""
Trains a subject specific model for the given subject
Parameters:
- subject: Integer in the Range 1 <= subject <= 9
- epochs: Number of epochs to train
- batch_size: Batch Size
- lr: Learning Rate
- silent: bool, if True, hide all output including the progress bar
- plot: bool, if True, generate plots
- kwargs: Remaining arguments passed to the EEGnet model
Returns: (model, metrics)
- model: t.nn.Module, trained model
- metrics: t.tensor, size=[1, 4], accuracy, precision, recall, f1
"""
# load the data
train_samples, train_labels = get_data(subject, training=True)
test_samples, test_labels = get_data(subject, training=False)
train_loader = as_data_loader(train_samples,
train_labels,
batch_size=batch_size)
# test_loader = as_data_loader(test_samples, test_labels, batch_size=test_labels.shape[0])
test_loader = as_data_loader(test_samples,
test_labels,
batch_size=batch_size)
# prepare quantization configuration
qconfig = tq.QConfig(
activation=tq.MinMaxObserver.with_args(dtype=t.quint8),
weight=tq.MinMaxObserver.with_args(dtype=t.qint8))
# prepare the model
model = EEGNetQuant(T=train_samples.shape[2], qconfig=qconfig, **kwargs)
model.initialize_params()
if t.cuda.is_available():
model = model.cuda()
# prepare the quantization
tq.prepare_qat(model, inplace=True)
# prepare loss function and optimizer
loss_function = t.nn.CrossEntropyLoss()
optimizer = t.optim.Adam(model.parameters(), lr=lr, eps=1e-7)
scheduler = None
# print the training setup
print_summary(model, optimizer, loss_function, scheduler)
# prepare progress bar
with tqdm(desc=f"Subject {subject}",
total=epochs,
leave=False,
disable=silent,
unit='epoch',
ascii=True) as pbar:
# Early stopping is not allowed in this mode, because the testing data cannot be used for
# training!
model, metrics, _, history = _train_net(subject,
model,
train_loader,
test_loader,
loss_function,
optimizer,
scheduler=scheduler,
epochs=epochs,
early_stopping=False,
plot=plot,
pbar=pbar)
# convert the model into a quantized model
model = model.cpu()
tq.convert(model, inplace=True)
metrics = get_metrics_from_model(model, test_loader)
if not silent:
print(f"Subject {subject}: accuracy = {metrics[0, 0]}")
return model, metrics, history
def _qat_swap_modules(
self, root: torch.nn.Module,
additional_qat_module_mapping: Dict[Callable, Callable]) -> None:
all_mappings = get_combined_dict(
get_default_qat_module_mappings(), additional_qat_module_mapping)
convert(root, mapping=all_mappings, inplace=True, remove_qconfig=False)
def _test_model_impl(self,
mode,
name,
model,
eager_quantizable_model,
check_with_eager=True,
diff_of_quant=None,
diff_from_eager=None):
if diff_of_quant is None or diff_from_eager is None:
diff_of_quant = {}
diff_from_eager = {}
if mode not in diff_of_quant or mode not in diff_from_eager:
diff_of_quant[mode] = {}
diff_from_eager[mode] = {}
input_tensor = torch.rand(1, 3, 224, 224)
input_tensor_inception = torch.rand(1, 3, 299, 299)
output_value = torch.randint(0, 1, (1, ))
# print('quantizing:', name, ' mode:', mode)
if name == 'inception_v3':
input_value = input_tensor_inception
else:
input_value = input_tensor
qconfig = default_qconfig if mode == 'static' else default_qat_qconfig
qconfig_dict = {'': qconfig}
graph_module = symbolic_trace(model)
# print('graph module:', graph_module.src)
script = torch.jit.script(graph_module)
# make sure graph module and script module are both runanble
original_out = graph_module(input_value)
is_not_tuple_out = not isinstance(original_out, tuple)
script_out = script(input_value)
self.assertEqual(
(original_out - script_out).abs().max(), 0,
'Reslut of original graph module and script module does not match')
# set to train just before quantization
if mode != 'static':
model.train()
graph_module = fuse_fx(graph_module)
prepared = prepare_fx(graph_module, qconfig_dict)
if mode == 'ddp':
mp.spawn(run_ddp,
args=(world_size, prepared),
nprocs=world_size,
join=True)
elif mode == 'qat':
assert prepared.training, 'prepared must be in training mode for qat'
optimizer = torch.optim.SGD(prepared.parameters(), lr=0.0001)
criterion = nn.CrossEntropyLoss()
train_one_epoch(prepared, criterion,
optimizer, [(input_value, output_value)],
torch.device('cpu'), 1)
else:
for i in range(10):
prepared(input_value)
# print('after observation root:', prepared.root)
qgraph = convert_fx(prepared)
# print('after quantization root:', qgraph.root)
# print('after quantization code:', qgraph.src)
qgraph.eval()
qgraph_script = torch.jit.script(qgraph)
# print('quantized and scripted:', qgraph_script.graph)
qgraph_out = qgraph(input_value)
qgraph_script = qgraph_script(input_value)
if is_not_tuple_out:
diff_of_quant[mode][name] = (original_out - qgraph_out).abs().max()
assert torch.allclose(qgraph_out,
qgraph_script), 'graph, scripted graph'
else:
print('tuple output')
if eager_quantizable_model is not None:
# comparing to eager mode quantization
qeager = eager_quantizable_model
ref_out = qeager(input_value)
qeager.qconfig = qconfig
if mode == 'static':
qeager.fuse_model()
prepare(qeager, inplace=True)
else:
qeager.train()
qeager.fuse_model()
prepare_qat(qeager, inplace=True)
# calibration
if mode == 'ddp':
mp.spawn(run_ddp,
args=(world_size, qeager),
nprocs=world_size,
join=True)
elif mode == 'qat':
assert qeager.training, 'qeager should be in training mode for qat'
optimizer = torch.optim.SGD(qeager.parameters(), lr=0.0001)
train_one_epoch(qeager, criterion, optimizer,
[(input_value, output_value)],
torch.device('cpu'), 1)
else:
for i in range(10):
qeager(input_value)
# print('ref after observation:', qeager)
convert(qeager, inplace=True)
qeager.eval()
# print('ref after quantization:', qeager)
qeager_out = qeager(input_value)
qeager_script = torch.jit.script(qeager)
qscript_out = qeager_script(input_value)
if is_not_tuple_out:
diff_from_eager[mode][name] = (qeager_out -
qgraph_out).abs().max()
if check_with_eager:
self.assertEqual(
diff_from_eager[mode][name], 0,
'Result of graph mode quantization and ' +
'eager mode quantization on model: ' + name +
' should match. Mode: ' + mode + ' diff:' +
str(diff_from_eager[mode][name]))
def __init__(
self,
config,
model,
length_prediction_model,
trg_vocab,
beam_size,
use_gold_length,
beam_ranking_algorithm,
quantize,
embed_quantize,
):
super().__init__()
length_prediction_model = length_prediction_model.create_eval_module()
if quantize:
self.model = torch.quantization.quantize_dynamic(
model,
{
torch.nn.Linear:
torch.quantization.per_channel_dynamic_qconfig
},
dtype=torch.qint8,
inplace=False,
)
# embedding quantization
if embed_quantize != EmbedQuantizeType.NONE:
# 8-bit embedding quantization
if embed_quantize == EmbedQuantizeType.BIT_8:
## identify nn.Embedding
for module in self.model.modules():
if isinstance(module, torch.nn.Embedding):
module.qconfig = float_qparams_weight_only_qconfig
prepare(self.model, inplace=True)
convert(self.model, inplace=True)
# 4-bit embedding quantization
elif embed_quantize == EmbedQuantizeType.BIT_4:
raise NotImplementedError(
"4bit embedding quantization not yet supported")
else:
raise NotImplementedError(
"Embedding Quantization should be either 8bit or 4bit")
self.length_prediction_model = torch.quantization.quantize_dynamic(
length_prediction_model,
{
torch.nn.Linear:
torch.quantization.per_channel_dynamic_qconfig
},
dtype=torch.qint8,
inplace=False,
)
else:
self.model = model
self.length_prediction_model = length_prediction_model
self.trg_vocab = ScriptVocabulary(
list(trg_vocab),
pad_idx=trg_vocab.get_pad_index(),
bos_idx=trg_vocab.get_bos_index(-1),
eos_idx=trg_vocab.get_eos_index(-1),
mask_idx=trg_vocab.get_mask_index(),
)
self.length_beam_size = beam_size
self.use_gold_length = use_gold_length
self.beam_ranking_algorithm = get_beam_ranking_function(
ranking_algorithm=beam_ranking_algorithm)
self.clip_target_length = config.clip_target_length
self.targetlen_cap = config.targetlen_cap
self.targetlen_a = config.targetlen_a
self.targetlen_b = config.targetlen_b
self.targetlen_c = config.targetlen_c
def _qat_swap_modules(self, root):
convert(root, mapping=get_qat_module_mappings(), inplace=True, remove_qconfig=False)
def convert(model, path):
state_dict = load_model(path)
model.load_state_dict(state_dict)
model.eval()
opt_model = quantization.convert(model)
return opt_model
def _qat_swap_modules(self, root):
convert(root,
mapping=DEFAULT_QAT_MODULE_MAPPING,
inplace=True,
remove_qconfig=False)
def _qat_swap_modules(self, root, additional_qat_module_mapping):
all_mappings = get_combined_dict(get_default_qat_module_mappings(),
additional_qat_module_mapping)
convert(root, mapping=all_mappings, inplace=True, remove_qconfig=False)
print('6nd shape before reshape{}'.format(x.shape))
x = self.pool(F.relu(self.conv6(x)))
if DEBUG:
print('6nd shape before reshape{}'.format(x.shape))
x = x.view(-1, 875)
if DEBUG:
print('7nd_shape after reshape{}'.format(x.shape))
x = F.relu(self.fc1(x))
if DEBUG:
print('8nd_shape {}'.format(x.shape))
x = F.relu(self.fc2(x))
if DEBUG:
print('9nd_shape {}'.format(x.shape))
x = F.relu(self.fc3(x))
if DEBUG:
print('10nd_shape {}'.format(x.shape))
x = self.fc4(x)
if DEBUG:
print('11nd_shape {}'.format(x.shape))
return x
warnings.filterwarnings('ignore')
device = 'cuda'
model = torch.load('last_cnn.pt', 'cuda')
modelq = quantization.convert(model)
example = torch.rand(1, 3, 512, 512)
ex = example.to(device)
traced_script_module = torch.jit.trace(modelq, ex)
traced_script_module.save('last_jit_model_moda.pt')
def main():
parser = argparse.ArgumentParser()
# Required parameters
parser.add_argument("--model_type",
default=None,
type=str,
required=True,
help="Model type selected in the list: " +
", ".join(MODEL_CLASSES.keys()))
parser.add_argument(
"--model_name_or_path",
default=None,
type=str,
required=True,
help="Path to pre-trained model or shortcut name selected in the list: "
+ ", ".join(ALL_MODELS))
parser.add_argument(
"--output_dir",
default=None,
type=str,
required=True,
help=
"The output directory where the model checkpoints and predictions will be written."
)
# Other parameters
parser.add_argument(
"--data_dir",
default=None,
type=str,
help=
"The input data dir. Should contain the .json files for the task. If not specified, will run with tensorflow_datasets."
)
parser.add_argument(
"--config_name",
default="",
type=str,
help="Pretrained config name or path if not the same as model_name")
parser.add_argument(
"--tokenizer_name",
default="",
type=str,
help="Pretrained tokenizer name or path if not the same as model_name")
parser.add_argument(
"--cache_dir",
default="",
type=str,
help=
"Where do you want to store the pre-trained models downloaded from s3")
parser.add_argument(
'--version_2_with_negative',
action='store_true',
help=
'If true, the SQuAD examples contain some that do not have an answer.')
parser.add_argument(
'--null_score_diff_threshold',
type=float,
default=0.0,
help=
"If null_score - best_non_null is greater than the threshold predict null."
)
parser.add_argument(
"--max_seq_length",
default=384,
type=int,
help=
"The maximum total input sequence length after WordPiece tokenization. Sequences "
"longer than this will be truncated, and sequences shorter than this will be padded."
)
parser.add_argument(
"--doc_stride",
default=128,
type=int,
help=
"When splitting up a long document into chunks, how much stride to take between chunks."
)
parser.add_argument(
"--max_query_length",
default=64,
type=int,
help=
"The maximum number of tokens for the question. Questions longer than this will "
"be truncated to this length.")
parser.add_argument("--do_train",
action='store_true',
help="Whether to run training.")
parser.add_argument("--do_eval",
action='store_true',
help="Whether to run eval on the dev set.")
parser.add_argument(
"--evaluate_during_training",
action='store_true',
help="Rul evaluation during training at each logging step.")
parser.add_argument(
"--do_lower_case",
action='store_true',
help="Set this flag if you are using an uncased model.")
parser.add_argument("--per_gpu_train_batch_size",
default=8,
type=int,
help="Batch size per GPU/CPU for training.")
parser.add_argument("--per_gpu_eval_batch_size",
default=8,
type=int,
help="Batch size per GPU/CPU for evaluation.")
parser.add_argument("--learning_rate",
default=5e-5,
type=float,
help="The initial learning rate for Adam.")
parser.add_argument(
'--gradient_accumulation_steps',
type=int,
default=1,
help=
"Number of updates steps to accumulate before performing a backward/update pass."
)
parser.add_argument("--weight_decay",
default=0.0,
type=float,
help="Weight decay if we apply some.")
parser.add_argument("--adam_epsilon",
default=1e-8,
type=float,
help="Epsilon for Adam optimizer.")
parser.add_argument("--max_grad_norm",
default=1.0,
type=float,
help="Max gradient norm.")
parser.add_argument("--num_train_epochs",
default=3.0,
type=float,
help="Total number of training epochs to perform.")
parser.add_argument(
"--max_steps",
default=-1,
type=int,
help=
"If > 0: set total number of training steps to perform. Override num_train_epochs."
)
parser.add_argument("--warmup_steps",
default=0,
type=int,
help="Linear warmup over warmup_steps.")
parser.add_argument(
"--n_best_size",
default=20,
type=int,
help=
"The total number of n-best predictions to generate in the nbest_predictions.json output file."
)
parser.add_argument(
"--max_answer_length",
default=30,
type=int,
help=
"The maximum length of an answer that can be generated. This is needed because the start "
"and end predictions are not conditioned on one another.")
parser.add_argument(
"--verbose_logging",
action='store_true',
help=
"If true, all of the warnings related to data processing will be printed. "
"A number of warnings are expected for a normal SQuAD evaluation.")
parser.add_argument('--logging_steps',
type=int,
default=50,
help="Log every X updates steps.")
parser.add_argument('--save_steps',
type=int,
default=50,
help="Save checkpoint every X updates steps.")
parser.add_argument(
"--eval_all_checkpoints",
action='store_true',
help=
"Evaluate all checkpoints starting with the same prefix as model_name ending and ending with step number"
)
parser.add_argument("--no_cuda",
action='store_true',
help="Whether not to use CUDA when available")
parser.add_argument('--overwrite_output_dir',
action='store_true',
help="Overwrite the content of the output directory")
parser.add_argument(
'--overwrite_cache',
action='store_true',
help="Overwrite the cached training and evaluation sets")
parser.add_argument('--seed',
type=int,
default=42,
help="random seed for initialization")
parser.add_argument("--local_rank",
type=int,
default=-1,
help="local_rank for distributed training on gpus")
parser.add_argument(
'--fp16',
action='store_true',
help=
"Whether to use 16-bit (mixed) precision (through NVIDIA apex) instead of 32-bit"
)
parser.add_argument(
'--fp16_opt_level',
type=str,
default='O1',
help=
"For fp16: Apex AMP optimization level selected in ['O0', 'O1', 'O2', and 'O3']."
"See details at https://nvidia.github.io/apex/amp.html")
parser.add_argument('--server_ip',
type=str,
default='',
help="Can be used for distant debugging.")
parser.add_argument('--server_port',
type=str,
default='',
help="Can be used for distant debugging.")
parser.add_argument("--do_calibration",
action='store_true',
help="Whether to do calibration.")
parser.add_argument("--do_int8_inference",
action='store_true',
help="Whether to run int8 inference.")
parser.add_argument("--do_fp32_inference",
action='store_true',
help="Whether to run fp32 inference.")
parser.add_argument("--mkldnn_eval",
action='store_true',
help="evaluation with MKLDNN")
parser.add_argument(
"--tune",
action='store_true',
help="run Low Precision Optimization Tool to tune int8 acc.")
parser.add_argument("--task_name",
default=None,
type=str,
required=True,
help="SQuAD task")
parser.add_argument("--warmup",
type=int,
default=5,
help="warmup for performance")
parser.add_argument('-i',
"--iter",
default=0,
type=int,
help='For accuracy measurement only.')
parser.add_argument('--benchmark',
dest='benchmark',
action='store_true',
help='run benchmark')
parser.add_argument('-r',
"--accuracy_only",
dest='accuracy_only',
action='store_true',
help='For accuracy measurement only.')
parser.add_argument(
"--tuned_checkpoint",
default='./',
type=str,
metavar='PATH',
help=
'path to checkpoint tuned by Low Precision Optimization Tool (default: ./)'
)
parser.add_argument('--int8',
dest='int8',
action='store_true',
help='run benchmark')
args = parser.parse_args()
args.predict_file = os.path.join(
args.output_dir, 'predictions_{}_{}.txt'.format(
list(filter(None, args.model_name_or_path.split('/'))).pop(),
str(args.max_seq_length)))
if os.path.exists(args.output_dir) and os.listdir(
args.output_dir
) and args.do_train and not args.overwrite_output_dir:
raise ValueError(
"Output directory ({}) already exists and is not empty. Use --overwrite_output_dir to overcome."
.format(args.output_dir))
mix_qkv = False
if args.do_calibration or args.do_int8_inference or args.tune:
mix_qkv = True
# Setup distant debugging if needed
if args.server_ip and args.server_port:
# Distant debugging - see https://code.visualstudio.com/docs/python/debugging#_attach-to-a-local-script
import ptvsd
print("Waiting for debugger attach")
ptvsd.enable_attach(address=(args.server_ip, args.server_port),
redirect_output=True)
ptvsd.wait_for_attach()
# Setup CUDA, GPU & distributed training
if args.local_rank == -1 or args.no_cuda:
device = torch.device("cuda" if torch.cuda.is_available()
and not args.no_cuda else "cpu")
args.n_gpu = torch.cuda.device_count()
else: # Initializes the distributed backend which will take care of sychronizing nodes/GPUs
torch.cuda.set_device(args.local_rank)
device = torch.device("cuda", args.local_rank)
torch.distributed.init_process_group(backend='nccl')
args.n_gpu = 1
args.device = device
# Setup logging
logging.basicConfig(
format='%(asctime)s - %(levelname)s - %(name)s - %(message)s',
datefmt='%m/%d/%Y %H:%M:%S',
level=logging.INFO if args.local_rank in [-1, 0] else logging.WARN)
logger.warning(
"Process rank: %s, device: %s, n_gpu: %s, distributed training: %s, 16-bits training: %s",
args.local_rank, device, args.n_gpu, bool(args.local_rank != -1),
args.fp16)
# Set seed
set_seed(args)
# Load pretrained model and tokenizer
if args.local_rank not in [-1, 0]:
torch.distributed.barrier(
) # Make sure only the first process in distributed training will download model & vocab
args.model_type = args.model_type.lower()
config_class, model_class, tokenizer_class = MODEL_CLASSES[args.model_type]
config = config_class.from_pretrained(
args.config_name if args.config_name else args.model_name_or_path,
cache_dir=args.cache_dir if args.cache_dir else None)
tokenizer = tokenizer_class.from_pretrained(
args.tokenizer_name
if args.tokenizer_name else args.model_name_or_path,
do_lower_case=args.do_lower_case,
cache_dir=args.cache_dir if args.cache_dir else None)
model = model_class.from_pretrained(
args.model_name_or_path,
from_tf=bool('.ckpt' in args.model_name_or_path),
config=config,
mix_qkv=mix_qkv,
cache_dir=args.cache_dir if args.cache_dir else None)
if args.local_rank == 0:
torch.distributed.barrier(
) # Make sure only the first process in distributed training will download model & vocab
model.to(args.device)
logger.info("Training/evaluation parameters %s", args)
# Before we do anything with models, we want to ensure that we get fp16 execution of torch.einsum if args.fp16 is set.
# Otherwise it'll default to "promote" mode, and we'll get fp32 operations. Note that running `--fp16_opt_level="O2"` will
# remove the need for this code, but it is still valid.
if args.fp16:
try:
import apex
apex.amp.register_half_function(torch, 'einsum')
except ImportError:
raise ImportError(
"Please install apex from https://www.github.com/nvidia/apex to use fp16 training."
)
# Training
if args.do_train:
train_dataset = load_and_cache_examples(args,
tokenizer,
evaluate=False,
output_examples=False)
global_step, tr_loss = train(args, train_dataset, model, tokenizer)
logger.info(" global_step = %s, average loss = %s", global_step,
tr_loss)
# Save the trained model and the tokenizer
if args.do_train and (args.local_rank == -1
or torch.distributed.get_rank() == 0):
# Create output directory if needed
if not os.path.exists(args.output_dir) and args.local_rank in [-1, 0]:
os.makedirs(args.output_dir)
logger.info("Saving model checkpoint to %s", args.output_dir)
# Save a trained model, configuration and tokenizer using `save_pretrained()`.
# They can then be reloaded using `from_pretrained()`
model_to_save = model.module if hasattr(
model,
'module') else model # Take care of distributed/parallel training
model_to_save.save_pretrained(args.output_dir)
tokenizer.save_pretrained(args.output_dir)
# Good practice: save your training arguments together with the trained model
torch.save(args, os.path.join(args.output_dir, 'training_args.bin'))
# Load a trained model and vocabulary that you have fine-tuned
model = model_class.from_pretrained(args.output_dir,
force_download=True,
mix_qkv=mix_qkv)
tokenizer = tokenizer_class.from_pretrained(
args.output_dir, do_lower_case=args.do_lower_case)
model.to(args.device)
# Evaluation - we can ask to evaluate all the checkpoints (sub-directories) in a directory
results = {}
if args.do_eval and args.local_rank in [-1, 0]:
checkpoints = [args.output_dir]
if args.eval_all_checkpoints:
checkpoints = list(
os.path.dirname(c) for c in sorted(
glob.glob(args.output_dir + '/**/' + WEIGHTS_NAME,
recursive=True)))
logging.getLogger("transformers.modeling_utils").setLevel(
logging.WARN) # Reduce model loading logs
logger.info("Evaluate the following checkpoints: %s", checkpoints)
for checkpoint in checkpoints:
# Reload the model
global_step = checkpoint.split(
'-')[-1] if len(checkpoints) > 1 else ""
if args.mkldnn_eval or args.do_fp32_inference:
model = model_class.from_pretrained(checkpoint,
force_download=True)
model.to(args.device)
# Evaluate
result, _ = evaluate(args,
model,
tokenizer,
prefix=global_step)
result = dict(
(k + ('_{}'.format(global_step) if global_step else ''), v)
for k, v in result.items())
results.update(result)
if args.tune:
def eval_func_for_lpot(model):
result, _ = evaluate(args, model, tokenizer)
for key in sorted(result.keys()):
logger.info(" %s = %s", key, str(result[key]))
bert_task_acc_keys = [
'best_f1', 'f1', 'mcc', 'spearmanr', 'acc'
]
for key in bert_task_acc_keys:
if key in result.keys():
logger.info("Finally Eval {}:{}".format(
key, result[key]))
acc = result[key]
break
return acc
model = model_class.from_pretrained(checkpoint,
force_download=True,
mix_qkv=True)
model.to(args.device)
dataset = load_and_cache_examples(args,
tokenizer,
evaluate=True,
output_examples=False)
args.eval_batch_size = args.per_gpu_eval_batch_size * max(
1, args.n_gpu)
eval_task = "squad"
from lpot import Quantization
quantizer = Quantization("./conf.yaml")
dataset = quantizer.dataset('bert',
dataset=dataset,
task=eval_task,
model_type=args.model_type)
test_dataloader = quantizer.dataloader(
dataset, batch_size=args.eval_batch_size)
quantizer(model, test_dataloader, eval_func=eval_func_for_lpot)
exit(0)
if args.benchmark or args.accuracy_only:
model = model_class.from_pretrained(checkpoint, mix_qkv=True)
model.to(args.device)
if args.int8:
from lpot.utils.pytorch import load
new_model = load(
os.path.abspath(
os.path.expanduser(args.tuned_checkpoint)), model)
else:
new_model = model
result, _ = evaluate(args,
new_model,
tokenizer,
prefix=global_step)
exit(0)
if args.do_calibration:
model = model_class.from_pretrained(checkpoint,
force_download=True,
mix_qkv=True)
model.to(args.device)
model.qconfig = default_per_channel_qconfig
propagate_qconfig_(model)
add_observer_(model)
# Evaluate
evaluate(args,
model,
tokenizer,
prefix=global_step,
calibration=True)
convert(model, inplace=True)
quantized_model_path = "squad" + str(
global_step) + "_quantized_model"
if not os.path.exists(quantized_model_path):
os.makedirs(quantized_model_path)
model.save_pretrained(quantized_model_path)
result, _ = evaluate(args,
model,
tokenizer,
prefix=global_step)
result = dict(
(k + ('_{}'.format(global_step) if global_step else ''), v)
for k, v in result.items())
results.update(result)
if args.do_int8_inference:
model = model_class.from_pretrained(checkpoint,
force_download=True,
mix_qkv=True)
model.to(args.device)
model.qconfig = default_per_channel_qconfig
propagate_qconfig_(model)
add_observer_(model)
convert(model, inplace=True)
quantized_model_path = "squad" + str(
global_step) + "_quantized_model"
if not os.path.exists(quantized_model_path):
logger.info("Please run calibration first!")
return
model_bin_file = os.path.join(quantized_model_path,
"pytorch_model.bin")
state_dict = torch.load(model_bin_file)
model.load_state_dict(state_dict)
print(model)
with torch.autograd.profiler.profile() as prof:
result, _ = evaluate(args,
model,
tokenizer,
prefix=global_step)
print(prof.key_averages().table(sort_by="cpu_time_total"))
result = dict(
(k + ('_{}'.format(global_step) if global_step else ''), v)
for k, v in result.items())
results.update(result)
logger.info("Results: {}".format(results))
return results
def _qat_swap_modules(self, root, additional_qat_module_mapping):
all_mappings = get_default_qat_module_mappings().copy()
for k, v in additional_qat_module_mapping.items():
all_mappings[k] = v
convert(root, mapping=all_mappings, inplace=True, remove_qconfig=False)
def convert_dynamic(module):
convert(module, DEFAULT_DYNAMIC_MODULE_MAPPING, inplace=True)
def main():
parser = argparse.ArgumentParser()
## Required parameters
parser.add_argument(
"--data_dir",
default=None,
type=str,
required=True,
help=
"The input data dir. Should contain the .tsv files (or other data files) for the task."
)
parser.add_argument("--model_type",
default=None,
type=str,
required=True,
help="Model type selected in the list: " +
", ".join(MODEL_CLASSES.keys()))
parser.add_argument(
"--model_name_or_path",
default=None,
type=str,
required=True,
help="Path to pre-trained model or shortcut name selected in the list;"
+ ", ".join(ALL_MODELS))
parser.add_argument(
"--task_name",
default=None,
type=str,
required=True,
help="The name of the task to train selected in the list: " +
", ".join(processors.keys()))
parser.add_argument(
"--output_dir",
default=None,
type=str,
required=True,
help=
"The output directory where the model predictions and checkpoints will be written."
)
## Other parameters
parser.add_argument(
"--config_name",
default="",
type=str,
help="Pretrained config name or path if not the same as model_name")
parser.add_argument(
"--tokenizer_name",
default="",
type=str,
help="Pretrained tokenizer name or path if not the same as model_name")
parser.add_argument(
"--cache_dir",
default="",
type=str,
help=
"Where do you want to store the pre-trained models downloaded from s3")
parser.add_argument(
"--max_seq_length",
default=128,
type=int,
help=
"The maximum total input sequence length after tokenization. Sequences longer "
"than this will be truncated, sequences shorter will be padded.")
parser.add_argument("--do_train",
action='store_true',
help="Whether to run training.")
parser.add_argument("--do_eval",
action='store_true',
help="Whether to run eval on the dev set.")
parser.add_argument(
"--evaluate_during_training",
action='store_true',
help="Rul evaluation during training at each logging step.")
parser.add_argument(
"--do_lower_case",
action='store_true',
help="Set this flag if you are using an uncased model.")
parser.add_argument("--per_gpu_train_batch_size",
default=8,
type=int,
help="Batch size per GPU/CPU for training.")
parser.add_argument("--per_gpu_eval_batch_size",
default=8,
type=int,
help="Batch size per GPU/CPU for evaluation.")
parser.add_argument(
'--gradient_accumulation_steps',
type=int,
default=1,
help=
"Number of updates steps to accumulate before performing a backward/update pass."
)
parser.add_argument("--learning_rate",
default=5e-5,
type=float,
help="The initial learning rate for Adam.")
parser.add_argument("--weight_decay",
default=0.0,
type=float,
help="Weight deay if we apply some.")
parser.add_argument("--adam_epsilon",
default=1e-8,
type=float,
help="Epsilon for Adam optimizer.")
parser.add_argument("--max_grad_norm",
default=1.0,
type=float,
help="Max gradient norm.")
parser.add_argument("--num_train_epochs",
default=3.0,
type=float,
help="Total number of training epochs to perform.")
parser.add_argument(
"--max_steps",
default=-1,
type=int,
help=
"If > 0: set total number of training steps to perform. Override num_train_epochs."
)
parser.add_argument("--warmup_steps",
default=0,
type=int,
help="Linear warmup over warmup_steps.")
parser.add_argument('--logging_steps',
type=int,
default=50,
help="Log every X updates steps.")
parser.add_argument('--save_steps',
type=int,
default=50,
help="Save checkpoint every X updates steps.")
parser.add_argument(
"--eval_all_checkpoints",
action='store_true',
help=
"Evaluate all checkpoints starting with the same prefix as model_name ending and ending with step number"
)
parser.add_argument("--no_cuda",
action='store_true',
help="Avoid using CUDA when available")
parser.add_argument("--mkldnn_eval",
action='store_true',
help="evaluation with MKLDNN")
parser.add_argument("--mkldnn_train",
action='store_true',
help="training with MKLDNN")
parser.add_argument('--overwrite_output_dir',
action='store_true',
help="Overwrite the content of the output directory")
parser.add_argument(
'--overwrite_cache',
action='store_true',
help="Overwrite the cached training and evaluation sets")
parser.add_argument('--seed',
type=int,
default=42,
help="random seed for initialization")
parser.add_argument(
'--fp16',
action='store_true',
help=
"Whether to use 16-bit (mixed) precision (through NVIDIA apex) instead of 32-bit"
)
parser.add_argument(
'--fp16_opt_level',
type=str,
default='O1',
help=
"For fp16: Apex AMP optimization level selected in ['O0', 'O1', 'O2', and 'O3']."
"See details at https://nvidia.github.io/apex/amp.html")
parser.add_argument("--local_rank",
type=int,
default=-1,
help="For distributed training: local_rank")
parser.add_argument('--server_ip',
type=str,
default='',
help="For distant debugging.")
parser.add_argument('--server_port',
type=str,
default='',
help="For distant debugging.")
parser.add_argument("--do_fp32_inference",
action='store_true',
help="Whether to run fp32 inference.")
parser.add_argument("--do_calibration",
action='store_true',
help="Whether to do calibration.")
parser.add_argument("--do_int8_inference",
action='store_true',
help="Whether to run int8 inference.")
parser.add_argument("--do_bf16",
action='store_true',
help="run bf16 evaluation / training.")
parser.add_argument("--tune",
action='store_true',
help="run ilit to tune int8 acc.")
parser.add_argument("--warmup",
type=int,
default=2,
help="warmup for performance")
args = parser.parse_args()
if os.path.exists(args.output_dir) and os.listdir(
args.output_dir
) and args.do_train and not args.overwrite_output_dir:
raise ValueError(
"Output directory ({}) already exists and is not empty. Use --overwrite_output_dir to overcome."
.format(args.output_dir))
# Setup distant debugging if needed
if args.server_ip and args.server_port:
# Distant debugging - see https://code.visualstudio.com/docs/python/debugging#_attach-to-a-local-script
import ptvsd
print("Waiting for debugger attach")
ptvsd.enable_attach(address=(args.server_ip, args.server_port),
redirect_output=True)
ptvsd.wait_for_attach()
# Setup CUDA, GPU & distributed training
if args.local_rank == -1 or args.no_cuda:
device = torch.device("cuda" if torch.cuda.is_available()
and not args.no_cuda else "cpu")
args.n_gpu = torch.cuda.device_count()
else: # Initializes the distributed backend which will take care of sychronizing nodes/GPUs
torch.cuda.set_device(args.local_rank)
device = torch.device("cuda", args.local_rank)
torch.distributed.init_process_group(backend='nccl')
args.n_gpu = 1
args.device = device
# Setup logging
logging.basicConfig(
format='%(asctime)s - %(levelname)s - %(name)s - %(message)s',
datefmt='%m/%d/%Y %H:%M:%S',
level=logging.INFO if args.local_rank in [-1, 0] else logging.WARN)
logger.warning(
"Process rank: %s, device: %s, n_gpu: %s, distributed training: %s, 16-bits training: %s",
args.local_rank, device, args.n_gpu, bool(args.local_rank != -1),
args.fp16)
# Set seed
set_seed(args)
# Prepare GLUE task
args.task_name = args.task_name.lower()
if args.task_name not in processors:
raise ValueError("Task not found: %s" % (args.task_name))
processor = processors[args.task_name]()
args.output_mode = output_modes[args.task_name]
label_list = processor.get_labels()
num_labels = len(label_list)
mix_qkv = False
if args.do_calibration or args.do_int8_inference or args.tune:
mix_qkv = True
# Load pretrained model and tokenizer
if args.local_rank not in [-1, 0]:
torch.distributed.barrier(
) # Make sure only the first process in distributed training will download model & vocab
args.model_type = args.model_type.lower()
config_class, model_class, tokenizer_class = MODEL_CLASSES[args.model_type]
config = config_class.from_pretrained(
args.config_name if args.config_name else args.model_name_or_path,
num_labels=num_labels,
finetuning_task=args.task_name,
cache_dir=args.cache_dir if args.cache_dir else None)
tokenizer = tokenizer_class.from_pretrained(
args.tokenizer_name
if args.tokenizer_name else args.model_name_or_path,
do_lower_case=args.do_lower_case,
cache_dir=args.cache_dir if args.cache_dir else None)
model = model_class.from_pretrained(
args.model_name_or_path,
from_tf=bool('.ckpt' in args.model_name_or_path),
config=config,
mix_qkv=mix_qkv,
bf16=args.do_bf16,
mkldnn_train=args.mkldnn_train,
cache_dir=args.cache_dir if args.cache_dir else None)
if args.local_rank == 0:
torch.distributed.barrier(
) # Make sure only the first process in distributed training will download model & vocab
model.to(args.device)
logger.info("Training/evaluation parameters %s", args)
# Training
if args.do_train:
train_dataset = load_and_cache_examples(args,
args.task_name,
tokenizer,
evaluate=False)
global_step, tr_loss = train(args, train_dataset, model, tokenizer)
logger.info(" global_step = %s, average loss = %s", global_step,
tr_loss)
# Saving best-practices: if you use defaults names for the model, you can reload it using from_pretrained()
if args.do_train and (args.local_rank == -1
or torch.distributed.get_rank() == 0):
# Create output directory if needed
if not os.path.exists(args.output_dir) and args.local_rank in [-1, 0]:
os.makedirs(args.output_dir)
logger.info("Saving model checkpoint to %s", args.output_dir)
# Save a trained model, configuration and tokenizer using `save_pretrained()`.
# They can then be reloaded using `from_pretrained()`
model_to_save = model.module if hasattr(
model,
'module') else model # Take care of distributed/parallel training
model_to_save.save_pretrained(args.output_dir)
tokenizer.save_pretrained(args.output_dir)
# Good practice: save your training arguments together with the trained model
torch.save(args, os.path.join(args.output_dir, 'training_args.bin'))
# Load a trained model and vocabulary that you have fine-tuned
model = model_class.from_pretrained(args.output_dir)
tokenizer = tokenizer_class.from_pretrained(args.output_dir)
model.to(args.device)
# Evaluation
results = {}
if args.do_eval and args.local_rank in [-1, 0]:
tokenizer = tokenizer_class.from_pretrained(
args.output_dir, do_lower_case=args.do_lower_case)
checkpoints = [args.output_dir]
if args.eval_all_checkpoints:
checkpoints = list(
os.path.dirname(c) for c in sorted(
glob.glob(args.output_dir + '/**/' + WEIGHTS_NAME,
recursive=True)))
logging.getLogger("transformers.modeling_utils").setLevel(
logging.WARN) # Reduce logging
logger.info("Evaluate the following checkpoints: %s", checkpoints)
for checkpoint in checkpoints:
global_step = checkpoint.split(
'-')[-1] if len(checkpoints) > 1 else ""
prefix = checkpoint.split(
'/')[-1] if checkpoint.find('checkpoint') != -1 else ""
logger.info("Evaluate:" + args.task_name)
if args.mkldnn_eval or args.do_fp32_inference or args.do_bf16:
model = model_class.from_pretrained(checkpoint)
model.to(args.device)
result = evaluate(args, model, tokenizer, prefix=prefix)
result = dict((k + '_{}'.format(global_step), v)
for k, v in result.items())
results.update(result)
if args.tune:
def eval_func_for_ilit(model):
result, perf = evaluate(args,
model,
tokenizer,
prefix=prefix)
bert_task_acc_keys = [
'acc_and_f1', 'f1', 'mcc', 'spearmanr', 'acc'
]
for key in bert_task_acc_keys:
if key in result.keys():
logger.info("Finally Eval {}:{}".format(
key, result[key]))
acc = result[key]
break
return acc
model = model_class.from_pretrained(checkpoint, mix_qkv=True)
model.to(args.device)
eval_task_names = (
"mnli", "mnli-mm") if args.task_name == "mnli" else (
args.task_name, )
for eval_task in eval_task_names:
eval_dataset = load_and_cache_examples(args,
eval_task,
tokenizer,
evaluate=True)
args.eval_batch_size = args.per_gpu_eval_batch_size * max(
1, args.n_gpu)
# multi-gpu eval
if args.n_gpu > 1:
model = torch.nn.DataParallel(model)
if args.mkldnn_eval:
from torch.utils import mkldnn as mkldnn_utils
model = mkldnn_utils.to_mkldnn(model)
print(model)
import ilit
tuner = ilit.Tuner("./conf.yaml")
if eval_task != "squad":
eval_task = 'classifier'
eval_dataset = tuner.dataset('bert',
dataset=eval_dataset,
task=eval_task)
test_dataloader = tuner.dataloader(
eval_dataset, batch_size=args.eval_batch_size)
tuner.tune(model,
test_dataloader,
eval_func=eval_func_for_ilit)
exit(0)
if args.do_calibration:
model = model_class.from_pretrained(checkpoint, mix_qkv=True)
model.to(args.device)
model.qconfig = default_per_channel_qconfig
fallback_layers = {}
if args.model_name_or_path == "bert-base-uncased" and args.task_name == "mrpc":
fallback_layers = {"bert.encoder.layer.9.output.dense."}
propagate_qconfig_(model)
fallback_layer(model,
layer_name="",
exculde_layers=fallback_layers)
add_observer_(model)
result, _ = evaluate(args,
model,
tokenizer,
prefix=global_step,
calibration=True)
convert(model, inplace=True)
quantized_model_path = args.task_name + "_quantized_model"
if not os.path.exists(quantized_model_path):
os.makedirs(quantized_model_path)
model.save_pretrained(quantized_model_path)
print(model)
result, _ = evaluate(args, model, tokenizer, prefix=prefix)
if args.do_int8_inference:
model = model_class.from_pretrained(checkpoint, mix_qkv=True)
model.to(args.device)
model.qconfig = default_per_channel_qconfig
fallback_layers = {}
if args.model_name_or_path == "bert-base-uncased" and args.task_name == "mrpc":
fallback_layers = {"bert.encoder.layer.9.output.dense."}
propagate_qconfig_(model)
fallback_layer(model,
layer_name="",
exculde_layers=fallback_layers)
add_observer_(model)
convert(model, inplace=True)
quantized_model_path = args.task_name + "_quantized_model"
if not os.path.exists(quantized_model_path):
logger.error(
"please do calibrantion befor run int8 inference")
return
prepare(model, inplace=True)
convert(model, inplace=True)
model_bin_file = os.path.join(quantized_model_path,
"pytorch_model.bin")
state_dict = torch.load(model_bin_file)
model.load_state_dict(state_dict)
result, _ = evaluate(args, model, tokenizer, prefix=prefix)
return results
