Fix intermittent qsim2 test failures

Signed-off-by: Michael Tuttle <quic_mtuttle@quicinc.com>

TrainingExtensions/torch/test/python/v2/quantsim/test_quantsim.py

@@ -39,9 +39,11 @@
 import os
 import json
 import pytest
+import random
+import numpy as np
 from aimet_common.quantsim_config.utils import get_path_for_per_channel_config
 from aimet_common.defs import QuantizationDataType
-from aimet_torch.quantsim import load_encodings_to_sim
+from aimet_torch.quantsim import load_encodings_to_sim, QuantScheme
 from aimet_torch.v2.quantsim import QuantizationSimModel
 from aimet_torch.v2.quantization.encoding_analyzer import PercentileEncodingAnalyzer
 from aimet_torch.v2.quantization.base import QuantizerBase
@@ -59,6 +61,12 @@ def encodings_are_close(quantizer_1: AffineQuantizerBase, quantizer_2: AffineQua
            and quantizer_1.bitwidth == quantizer_2.bitwidth \
            and quantizer_1.symmetric == quantizer_2.symmetric
 
+@pytest.fixture(autouse=True)
+def set_seed():
+    random.seed(0)
+    torch.manual_seed(0)
+    np.random.seed(0)
+
 
 class ConcatModel(torch.nn.Module):
 
@@ -103,7 +111,7 @@ def forward_pass(model, args):
     def test_set_and_freeze_param_encodings(self, config_file):
         model = test_models.BasicConv2d(kernel_size=3)
         dummy_input = torch.rand(1, 64, 16, 16)
-        sim = QuantizationSimModel(model, dummy_input, config_file=config_file)
+        sim = QuantizationSimModel(model, dummy_input, quant_scheme=QuantScheme.post_training_tf, config_file=config_file)
         sim.compute_encodings(lambda model, _: model(dummy_input), None)
 
         with tempfile.TemporaryDirectory() as temp_dir:
@@ -149,7 +157,7 @@ def test_set_and_freeze_param_encodings(self, config_file):
     def test_load_and_freeze_encodings(self, config_file):
         model = test_models.TinyModel()
         dummy_input = torch.rand(1, 3, 32, 32)
-        sim = QuantizationSimModel(model, dummy_input, config_file=config_file)
+        sim = QuantizationSimModel(model, dummy_input, quant_scheme=QuantScheme.post_training_tf, config_file=config_file)
         sim.compute_encodings(lambda model, _: model(dummy_input), None)
 
         with tempfile.TemporaryDirectory() as temp_dir:
@@ -208,7 +216,7 @@ def test_load_and_freeze_with_partial_encodings(self):
             "param_encodings": {"conv1.weight": [sample_encoding]}
         }
 
-        sim = QuantizationSimModel(model, dummy_input)
+        sim = QuantizationSimModel(model, dummy_input, quant_scheme=QuantScheme.post_training_tf)
         all_quantizers = [q for q in sim.model.modules() if isinstance(q, QuantizerBase)]
         sim.load_and_freeze_encodings(partial_encodings)