add pruner and quanter for theseus

ppcls/arch/__init__.py

@@ -26,15 +26,20 @@ from .utils import *
 from ppcls.arch.backbone.base.theseus_layer import TheseusLayer
 from ppcls.utils import logger
 from ppcls.utils.save_load import load_dygraph_pretrain
+from ppcls.engine.slim import prune_model, quantize_model
+
 
 __all__ = ["build_model", "RecModel", "DistillationModel"]
 
 
 def build_model(config):
-    config = copy.deepcopy(config)
-    model_type = config.pop("name")
+    arch_config = copy.deepcopy(config["Arch"])
+    model_type = arch_config.pop("name")
     mod = importlib.import_module(__name__)
-    arch = getattr(mod, model_type)(**config)
+    arch = getattr(mod, model_type)(**arch_config)
+    if isinstance(arch, TheseusLayer):
+        prune_model(config, arch)
+        quantize_model(config, arch)
     return arch
 
 
@@ -51,7 +56,7 @@ def apply_to_static(config, model):
     return model
 
 
-class RecModel(nn.Layer):
+class RecModel(TheseusLayer):
     def __init__(self, **config):
         super().__init__()
         backbone_config = config["Backbone"]

ppcls/arch/backbone/base/theseus_layer.py

@@ -16,6 +16,8 @@ class TheseusLayer(nn.Layer):
         super(TheseusLayer, self).__init__()
         self.res_dict = {}
         self.res_name = self.full_name()
+        self.pruner = None
+        self.quanter = None
 
     # stop doesn't work when stop layer has a parallel branch.
     def stop_after(self, stop_layer_name: str):

ppcls/engine/engine.py

@@ -44,7 +44,6 @@ from ppcls.data import create_operators
 from ppcls.engine.train import train_epoch
 from ppcls.engine import evaluation
 from ppcls.arch.gears.identity_head import IdentityHead
-from ppcls.engine.slim import get_pruner, get_quaner
 
 
 class Engine(object):
@@ -186,14 +185,12 @@ class Engine(object):
             self.eval_metric_func = None
 
         # build model
-        self.model = build_model(self.config["Arch"])
+        self.model = build_model(self.config)
+        self.quanted = self.config.get("Slim", {}).get("quant", False)
+        self.pruned = self.config.get("Slim", {}).get("prune", False)
         # set @to_static for benchmark, skip this by default.
         apply_to_static(self.config, self.model)
 
-        # for slim
-        self.pruner = get_pruner(self.config, self.model)
-        self.quanter = get_quaner(self.config, self.model)
-
         # load_pretrain
         if self.config["Global"]["pretrained_model"] is not None:
             if self.config["Global"]["pretrained_model"].startswith("http"):
@@ -371,8 +368,8 @@ class Engine(object):
         model.eval()
         save_path = os.path.join(self.config["Global"]["save_inference_dir"],
                                  "inference")
-        if self.quanter:
-            self.quanter.save_quantized_model(
+        if self.quanted:
+            model.quanter.save_quantized_model(
                 model.base_model,
                 save_path,
                 input_spec=[

ppcls/engine/slim/__init__.py

@@ -12,5 +12,5 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-from ppcls.engine.slim.prune import get_pruner
-from ppcls.engine.slim.quant import get_quaner
+from ppcls.engine.slim.prune import prune_model
+from ppcls.engine.slim.quant import quantize_model

ppcls/engine/slim/prune.py

@@ -17,7 +17,7 @@ import paddle
 from ppcls.utils import logger
 
 
-def get_pruner(config, model):
+def prune_model(config, model):
     if config.get("Slim", False) and config["Slim"].get("prune", False):
         import paddleslim
         prune_method_name = config["Slim"]["prune"]["name"].lower()
@@ -25,21 +25,20 @@ def get_pruner(config, model):
             "fpgm", "l1_norm"
         ], "The prune methods only support 'fpgm' and 'l1_norm'"
         if prune_method_name == "fpgm":
-            pruner = paddleslim.dygraph.FPGMFilterPruner(
+            model.pruner = paddleslim.dygraph.FPGMFilterPruner(
                 model, [1] + config["Global"]["image_shape"])
         else:
-            pruner = paddleslim.dygraph.L1NormFilterPruner(
+            model.pruner = paddleslim.dygraph.L1NormFilterPruner(
                 model, [1] + config["Global"]["image_shape"])
 
         # prune model
-        _prune_model(pruner, config, model)
+        _prune_model(config, model)
     else:
-        pruner = None
-
-    return pruner
+        model.pruner = None
 
 
-def _prune_model(pruner, config, model):
+
+def _prune_model(config, model):
     from paddleslim.analysis import dygraph_flops as flops
     logger.info("FLOPs before pruning: {}GFLOPs".format(
         flops(model, [1] + config["Global"]["image_shape"]) / 1e9))
@@ -53,7 +52,7 @@ def _prune_model(pruner, config, model):
     ratios = {}
     for param in params:
         ratios[param] = config["Slim"]["prune"]["pruned_ratio"]
-    plan = pruner.prune_vars(ratios, [0])
+    plan = model.pruner.prune_vars(ratios, [0])
 
     logger.info("FLOPs after pruning: {}GFLOPs; pruned ratio: {}".format(
         flops(model, [1] + config["Global"]["image_shape"]) / 1e9,

ppcls/engine/slim/quant.py

@@ -40,16 +40,16 @@ QUANT_CONFIG = {
 }
 
 
-def get_quaner(config, model):
+def quantize_model(config, model):
     if config.get("Slim", False) and config["Slim"].get("quant", False):
         from paddleslim.dygraph.quant import QAT
         assert config["Slim"]["quant"]["name"].lower(
         ) == 'pact', 'Only PACT quantization method is supported now'
         QUANT_CONFIG["activation_preprocess_type"] = "PACT"
-        quanter = QAT(config=QUANT_CONFIG)
-        quanter.quantize(model)
+        model.quanted = QAT(config=QUANT_CONFIG)
+        model.quanted.quantize_model(model)
         logger.info("QAT model summary:")
         paddle.summary(model, (1, 3, 224, 224))
     else:
-        quanter = None
-    return quanter
+        model.quanted = None
+    return model.quanted

ppcls/static/program.py

@@ -259,7 +259,7 @@ def build(config,
             # data_format should be assigned in arch-dict
             input_image_channel = config["Global"]["image_shape"][
                 0]  # default as [3, 224, 224]
-            model = build_model(config["Arch"])
+            model = build_model(config)
             out = model(feeds["data"])
             # end of build model