updates for v0.4.0 release preparation

Signed-off-by: Ranganath Krishnan <[email protected]>

LICENSE

@@ -1,6 +1,6 @@
 BSD 3-Clause License
 
-Copyright (c) 2021, Intel Labs
+Copyright (c) 2023, Intel Labs
 All rights reserved.
 
 Redistribution and use in source and binary forms, with or without

README.md

@@ -7,7 +7,7 @@ A library for Bayesian neural network layers and uncertainty estimation in Deep
 
 [![python](https://img.shields.io/badge/python-3.7%2B-blue)](https://github.com/IntelLabs/bayesian-torch)
 [![pytorch](https://img.shields.io/badge/pytorch-1.7.0%2B-orange)](https://github.com/IntelLabs/bayesian-torch)
-[![version](https://img.shields.io/badge/release-0.3.0-green)](https://github.com/IntelLabs/bayesian-torch/releases)
+[![version](https://img.shields.io/badge/release-0.4.0-green)](https://github.com/IntelLabs/bayesian-torch/releases)
 [![license](https://img.shields.io/badge/license-BSD%203--Clause-blue)](https://github.com/IntelLabs/bayesian-torch/blob/main/LICENSE)
 [![Downloads](https://static.pepy.tech/badge/bayesian-torch)](https://pepy.tech/project/bayesian-torch)
 <h4 align="center">

bayesian_torch/ao/nn/quantized/modules/quantize_conv_variational.py

@@ -1,4 +1,4 @@
-# Copyright (C) 2021 Intel Labs
+# Copyright (C) 2023 Intel Labs
 #
 # BSD-3-Clause License
 #
@@ -1425,4 +1425,4 @@ def forward(self, input, enable_int8_compute=True, normal_scale=6/255, default_s
         if return_kl:
             return out, 0 # disable kl divergence computing
 
-        return out
+        return out

bayesian_torch/ao/nn/quantized/modules/quantize_linear_variational.py

@@ -1,4 +1,4 @@
-# Copyright (C) 2021 Intel Labs
+# Copyright (C) 2023 Intel Labs
 #
 # BSD-3-Clause License
 #

bayesian_torch/ao/nn/quantized/modules/quantized_conv_flipout.py

@@ -1,4 +1,4 @@
-# Copyright (C) 2021 Intel Labs
+# Copyright (C) 2023 Intel Labs
 #
 # BSD-3-Clause License
 #
@@ -1300,4 +1300,4 @@ def forward(self, x, normal_scale=6/255, default_scale=0.1, default_zero_point=1
         if return_kl:
             return out, 0
 
-        return out
+        return out

bayesian_torch/ao/nn/quantized/modules/quantized_linear_flipout.py

@@ -1,4 +1,4 @@
-# Copyright (C) 2021 Intel Labs
+# Copyright (C) 2023 Intel Labs
 #
 # BSD-3-Clause License
 #

bayesian_torch/ao/quantization/quantize.py

@@ -1,4 +1,4 @@
-# Copyright (C) 2021 Intel Labs
+# Copyright (C) 2023 Intel Labs
 #
 # BSD-3-Clause License
 #

bayesian_torch/layers/base_variational_layer.py

@@ -1,4 +1,4 @@
-# Copyright (C) 2021 Intel Labs
+# Copyright (C) 2023 Intel Labs
 #
 # BSD-3-Clause License
 #

bayesian_torch/layers/flipout_layers/conv_flipout.py

@@ -1,4 +1,4 @@
-# Copyright (C) 2021 Intel Labs
+# Copyright (C) 2023 Intel Labs
 #
 # BSD-3-Clause License
 #

bayesian_torch/layers/flipout_layers/linear_flipout.py

@@ -1,4 +1,4 @@
-# Copyright (C) 2021 Intel Labs
+# Copyright (C) 2023 Intel Labs
 #
 # BSD-3-Clause License
 #

bayesian_torch/layers/flipout_layers/quantized_conv_flipout.py

@@ -1,4 +1,4 @@
-# Copyright (C) 2021 Intel Labs
+# Copyright (C) 2023 Intel Labs
 #
 # BSD-3-Clause License
 #
@@ -1348,4 +1348,4 @@ def forward(self, x, normal_scale=6/255, default_scale=0.1, default_zero_point=1
         if return_kl:
             return out, 0
 
-        return out
+        return out

bayesian_torch/layers/flipout_layers/quantized_linear_flipout.py

@@ -1,4 +1,4 @@
-# Copyright (C) 2021 Intel Labs
+# Copyright (C) 2023 Intel Labs
 #
 # BSD-3-Clause License
 #

bayesian_torch/layers/flipout_layers/rnn_flipout.py

@@ -1,4 +1,4 @@
-# Copyright (C) 2021 Intel Labs
+# Copyright (C) 2023 Intel Labs
 #
 # BSD-3-Clause License
 #

bayesian_torch/layers/variational_layers/conv_variational.py

@@ -1,4 +1,4 @@
-# Copyright (C) 2021 Intel Labs
+# Copyright (C) 2023 Intel Labs
 #
 # BSD-3-Clause License
 #

bayesian_torch/layers/variational_layers/linear_variational.py

@@ -1,4 +1,4 @@
-# Copyright (C) 2021 Intel Labs
+# Copyright (C) 2023 Intel Labs
 #
 # BSD-3-Clause License
 #

bayesian_torch/layers/variational_layers/quantize_conv_variational.py

@@ -1,4 +1,4 @@
-# Copyright (C) 2021 Intel Labs
+# Copyright (C) 2023 Intel Labs
 #
 # BSD-3-Clause License
 #
@@ -1489,4 +1489,4 @@ def forward(self, input, enable_int8_compute=True, normal_scale=6/255, default_s
         if return_kl:
             return out, 0 # disable kl divergence computing
 
-        return out
+        return out

bayesian_torch/layers/variational_layers/quantize_linear_variational.py

@@ -1,4 +1,4 @@
-# Copyright (C) 2021 Intel Labs
+# Copyright (C) 2023 Intel Labs
 #
 # BSD-3-Clause License
 #

bayesian_torch/layers/variational_layers/rnn_variational.py

@@ -1,4 +1,4 @@
-# Copyright (C) 2021 Intel Labs
+# Copyright (C) 2023 Intel Labs
 #
 # BSD-3-Clause License
 #

bayesian_torch/models/bnn_to_qbnn.py

@@ -1,4 +1,4 @@
-# Copyright (C) 2021 Intel Labs
+# Copyright (C) 2023 Intel Labs
 #
 # BSD-3-Clause License
 #
@@ -256,4 +256,4 @@ def forward(self, x):
     print(m)
     if input.dtype!=torch.quint8:
         input = torch.quantize_per_tensor(input, 0.1, 128, torch.quint8)
-    print(m(input))
+    print(m(input))

bayesian_torch/models/dnn_to_bnn.py

@@ -1,4 +1,4 @@
-# Copyright (C) 2021 Intel Labs
+# Copyright (C) 2023 Intel Labs
 #
 # BSD-3-Clause License
 #

bayesian_torch/scripts/quantize_bayesian_cifar.sh

@@ -0,0 +1,9 @@
+#!/bin/bash
+
+model=resnet20
+mode='ptq'
+batch_size=1000
+num_monte_carlo=20
+checkpoint_file=./checkpoint/bayesian/bayesian_resnet20_cifar.pth
+
+python examples/main_bayesian_cifar_dnn2bnn.py --arch=$model --mode=$mode --batch-size=$batch_size --num_monte_carlo=$num_monte_carlo --model-checkpoint=$checkpoint_file

bayesian_torch/utils/avuc_loss.py

@@ -1,4 +1,4 @@
-# Copyright (C) 2021 Intel Corporation
+# Copyright (C) 2023 Intel Corporation
 #
 # BSD-3-Clause License
 #

bayesian_torch/utils/util.py

@@ -1,4 +1,4 @@
-# Copyright (C) 2021 Intel Labs 
+# Copyright (C) 2023 Intel Labs 
 #
 # BSD-3-Clause License
 #

setup.py

@@ -11,7 +11,7 @@
 setup(
     name = "bayesian-torch",
     packages = find_packages(),
-    version = "0.3.0",
+    version = "0.4.0",
     description = "A library for Bayesian neural network layers and uncertainty estimation in Deep Learning",
     author = "Intel Labs",
     author_email = "[email protected]",