[quantization] Disable Conv1d for GPTQFPI/ GPTQ (#409)

This PR disables `conv1d` nodes for further investigation.

TICO-DCO-1.0-Signed-off-by: s.malakhov <s.malakhov@partner.samsung.com>

Author: stamalakhov

tico/quantization/algorithm/fpi_gptq/quantizer.py

@@ -76,9 +76,7 @@ def convert(self, model):
             )
         ):
             # 1) Identify quantizable submodules within the layer
-            full = find_layers(
-                layer, layers=[torch.nn.Linear, torch.nn.Conv2d, torch.nn.Conv1d]
-            )
+            full = find_layers(layer, layers=[torch.nn.Linear, torch.nn.Conv2d])
             # filter out depthwise convolutions and alike
             full = {
                 key: full[key]

tico/quantization/algorithm/gptq/quantizer.py

@@ -193,9 +193,7 @@ def convert(self, model):
             )
         ):
             # 1) Identify quantizable submodules within the layer
-            full = find_layers(
-                layer, layers=[torch.nn.Linear, torch.nn.Conv2d, torch.nn.Conv1d]
-            )
+            full = find_layers(layer, layers=[torch.nn.Linear, torch.nn.Conv2d])
             # filter out depthwise convolutions and alike
             full = {
                 key: full[key]