Comments

Makefile

@@ -126,7 +126,6 @@ install-torch-test:
 	pip install -e .
 	pip install "git+https://github.com/openvinotoolkit/open_model_zoo.git@37f60eb#egg=accuracy_checker&subdirectory=tools/accuracy_checker"
 	pip install -r tests/torch/requirements.txt
-	pip install -r tests/torch/fx/requirements.txt
 	pip install -r tests/cross_fw/install/requirements.txt
 	pip install -r tests/cross_fw/examples/requirements.txt
 

nncf/experimental/torch/fx/model_transformer.py

@@ -10,8 +10,6 @@
 # limitations under the License.
 
 from collections import defaultdict
-
-# from functools import partial
 from typing import Callable, List, Union
 
 import torch
@@ -63,9 +61,9 @@ def transform(self, transformation_layout: PTTransformationLayout) -> torch.fx.G
             if transformations:
                 model = transformation_fn(model, transformations)
 
-        # Do not eliminate dead code as
-        # the dead code is computing statistics :)
-        # model.graph.eliminate_dead_code()
+        # Do not use model.graph.eliminate_dead_code()
+        # because the computational statistics code
+        # is interpolated as dead code.
         model.recompile()
         return model
 

nncf/experimental/torch/fx/quantization/quantize_model.py

@@ -94,11 +94,21 @@ def quantize_impl(
         advanced_parameters=advanced_parameters,
     )
 
+    # BatchNorm operations have 3 output ports,
+    # to make it easier for alorithms to work
+    # with the target graph BatchNorm operations
+    # are being fused
     _fuse_conv_bn_(copied_model)
-    # BN fuses to conv bias, conv+bias joined op
-    # needs to be splited for nncf
+
+    # To make it easier for bias correction algorithms,
+    # biases are being separated by the followng calls.
     separate_linear_and_bias(copied_model)
     separate_conv_and_bias(copied_model)
+
+    # View requires at least one dimension spans
+    # across two contiguous subspaces and reshape is not.
+    # To prevent error during statistics collection
+    # all view operation are translated to reshape.
     view_to_reshape(copied_model)
 
     nncf_graph = NNCFGraphFactory.create(copied_model)

nncf/experimental/torch/fx/transformations.py

@@ -140,7 +140,7 @@ def insert_one_qdq_before_node(model: torch.fx.GraphModule, target_node: torch.f
     # 1. extract information for inserting q/dq node from activation_post_process
     node_type = "call_function"
     quantize_op: Optional[Callable] = None
-    # scale, zero_point = activation_post_process.calculate_qparams()  # type: ignore[attr-defined, operator]
+
     dtype = torch.int8 if quantizer.quant_min < 0 else torch.uint8
     if quantizer.is_per_channel:
         qparams = {

tests/torch/requirements.txt

@@ -19,3 +19,8 @@ datasets==2.14.7
 evaluate==0.3.0
 openvino
 timm==0.9.2
+
+
+# Required for torch/fx tests
+torchvision
+fastdownload==0.0.7