Fixing hyperlinks

PyTorch/LanguageModeling/BERT/README.md

@@ -97,7 +97,7 @@ Other publicly available implementations of BERT include:
 1. [NVIDIA TensorFlow](https://github.com/NVIDIA/DeepLearningExamples/tree/master/TensorFlow/LanguageModeling/BERT)
 2. [Hugging Face](https://github.com/huggingface/pytorch-pretrained-BERT)
 3. [codertimo](https://github.com/codertimo/BERT-pytorch)
-4. [gluon-nlp](https://github.com/dmlc/gluon-nlp/tree/master/scripts/bert)
+4. [gluon-nlp](https://github.com/dmlc/gluon-nlp/tree/v0.10.x/scripts/bert)
 5. [Google's implementation](https://github.com/google-research/bert)
     
 This model trains with mixed precision Tensor Cores on Volta and provides a push-button solution to pretraining on a corpus of choice. As a result, researchers can get results 4x faster than training without Tensor Cores. This model is tested against each NGC monthly container release to ensure consistent accuracy and performance over time.

TensorFlow/Classification/ConvNets/resnet50v1.5/README.md

@@ -200,7 +200,7 @@ cd DeepLearningExamples/TensorFlow/Classification/ConvNets
 2. Download and preprocess the dataset.
 The ResNet50 v1.5 script operates on ImageNet 1k, a widely popular image classification dataset from the ILSVRC challenge.
 
-To download and preprocess the dataset, use the [Generate ImageNet for TensorFlow](https://github.com/tensorflow/models/blob/master/research/inception/inception/data/download_and_preprocess_imagenet.sh) script. The dataset will be downloaded to a directory specified as the first parameter of the script.
+To download and preprocess the dataset, use the [Generate ImageNet for TensorFlow](https://github.com/tensorflow/models/blob/archive/research/inception/inception/data/download_and_preprocess_imagenet.sh) script. The dataset will be downloaded to a directory specified as the first parameter of the script.
 
 3. Build the ResNet-50 v1.5 TensorFlow NGC container.
 ```bash

TensorFlow/Classification/ConvNets/resnext101-32x4d/README.md

@@ -209,7 +209,7 @@ cd DeepLearningExamples/TensorFlow/Classification/ConvNets
 2. Download and preprocess the dataset.
 The ResNext101-32x4d script operates on ImageNet 1k, a widely popular image classification dataset from the ILSVRC challenge.
 
-To download and preprocess the dataset, use the [Generate ImageNet for TensorFlow](https://github.com/tensorflow/models/blob/master/research/inception/inception/data/download_and_preprocess_imagenet.sh) script. The dataset will be downloaded to a directory specified as the first parameter of the script.
+To download and preprocess the dataset, use the [Generate ImageNet for TensorFlow](https://github.com/tensorflow/models/blob/archive/research/inception/inception/data/download_and_preprocess_imagenet.sh) script. The dataset will be downloaded to a directory specified as the first parameter of the script.
 
 3. Build the ResNext101-32x4d TensorFlow NGC container.
 ```bash

TensorFlow/Classification/ConvNets/se-resnext101-32x4d/README.md

@@ -204,7 +204,7 @@ cd DeepLearningExamples/TensorFlow/Classification/ConvNets
 2. Download and preprocess the dataset.
 The SE-ResNext101-32x4d script operates on ImageNet 1k, a widely popular image classification dataset from the ILSVRC challenge.
 
-To download and preprocess the dataset, use the [Generate ImageNet for TensorFlow](https://github.com/tensorflow/models/blob/master/research/inception/inception/data/download_and_preprocess_imagenet.sh) script. The dataset will be downloaded to a directory specified as the first parameter of the script.
+To download and preprocess the dataset, use the [Generate ImageNet for TensorFlow](https://github.com/tensorflow/models/blob/archive/research/inception/inception/data/download_and_preprocess_imagenet.sh) script. The dataset will be downloaded to a directory specified as the first parameter of the script.
 
 3. Build the SE-ResNext101-32x4d TensorFlow NGC container.
 ```bash

TensorFlow/LanguageModeling/BERT/README.md

@@ -85,7 +85,7 @@ Other publicly available implementations of BERT include:
 1. [NVIDIA PyTorch](https://github.com/NVIDIA/DeepLearningExamples/tree/master/PyTorch/LanguageModeling/BERT)
 2. [Hugging Face](https://github.com/huggingface/pytorch-pretrained-BERT)
 3. [codertimo](https://github.com/codertimo/BERT-pytorch)
-4. [gluon-nlp](https://github.com/dmlc/gluon-nlp/tree/master/scripts/bert)
+4. [gluon-nlp](https://github.com/dmlc/gluon-nlp/tree/v0.10.x/scripts/bert)
 5. [Google's official implementation](https://github.com/google-research/bert)
 
 This model is trained with mixed precision using Tensor Cores on NVIDIA Volta, Ampere and Turing GPUs. Therefore, researchers can get results up to 4x faster than training without Tensor Cores, while experiencing the benefits of mixed precision training. This model is tested against each NGC monthly container release to ensure consistent accuracy and performance over time.

TensorFlow/Segmentation/UNet_Industrial/README.md

@@ -160,11 +160,11 @@ if params.use_amp:
  
 #### Enabling TF32
 
-TensorFloat-32 (TF32) is the new math mode in [NVIDIA A100](#https://www.nvidia.com/en-us/data-center/a100/) GPUs for handling the matrix math also called tensor operations. TF32 running on Tensor Cores in A100 GPUs can provide up to 10x speedups compared to single-precision floating-point math (FP32) on Volta GPUs. 
+TensorFloat-32 (TF32) is the new math mode in [NVIDIA A100](https://www.nvidia.com/en-us/data-center/a100/) GPUs for handling the matrix math also called tensor operations. TF32 running on Tensor Cores in A100 GPUs can provide up to 10x speedups compared to single-precision floating-point math (FP32) on Volta GPUs. 
 
 TF32 Tensor Cores can speed up networks using FP32, typically with no loss of accuracy. It is more robust than FP16 for models which require high dynamic range for weights or activations.
 
-For more information, refer to the [TensorFloat-32 in the A100 GPU Accelerates AI Training, HPC up to 20x](#https://blogs.nvidia.com/blog/2020/05/14/tensorfloat-32-precision-format/) blog post.
+For more information, refer to the [TensorFloat-32 in the A100 GPU Accelerates AI Training, HPC up to 20x](https://blogs.nvidia.com/blog/2020/05/14/tensorfloat-32-precision-format/) blog post.
 
 TF32 is supported in the NVIDIA Ampere GPU architecture and is enabled by default.
 

TensorFlow/Segmentation/VNet/README.md

@@ -125,11 +125,11 @@ if params['use_amp']:
 
 #### Enabling TF32
 
-TensorFloat-32 (TF32) is the new math mode in [NVIDIA A100](#https://www.nvidia.com/en-us/data-center/a100/) GPUs for handling the matrix math also called tensor operations. TF32 running on Tensor Cores in A100 GPUs can provide up to 10x speedups compared to single-precision floating-point math (FP32) on Volta GPUs. 
+TensorFloat-32 (TF32) is the new math mode in [NVIDIA A100](https://www.nvidia.com/en-us/data-center/a100/) GPUs for handling the matrix math also called tensor operations. TF32 running on Tensor Cores in A100 GPUs can provide up to 10x speedups compared to single-precision floating-point math (FP32) on Volta GPUs. 
 
 TF32 Tensor Cores can speed up networks using FP32, typically with no loss of accuracy. It is more robust than FP16 for models which require high dynamic range for weights or activations.
 
-For more information, refer to the [TensorFloat-32 in the A100 GPU Accelerates AI Training, HPC up to 20x](#https://blogs.nvidia.com/blog/2020/05/14/tensorfloat-32-precision-format/) blog post.
+For more information, refer to the [TensorFloat-32 in the A100 GPU Accelerates AI Training, HPC up to 20x](https://blogs.nvidia.com/blog/2020/05/14/tensorfloat-32-precision-format/) blog post.
 
 TF32 is supported in the NVIDIA Ampere GPU architecture and is enabled by default.
 

TensorFlow2/LanguageModeling/ELECTRA/README.md

@@ -142,11 +142,11 @@ To enable mixed precision, you can simply add the `--amp` to the command-line us
  
 #### Enabling TF32
  
-TensorFloat-32 (TF32) is the new math mode in [NVIDIA A100](#https://www.nvidia.com/en-us/data-center/a100/) GPUs for handling the matrix math also called tensor operations. TF32 running on Tensor Cores in A100 GPUs can provide up to 10x speedups compared to single-precision floating-point math (FP32) on Volta GPUs. 
+TensorFloat-32 (TF32) is the new math mode in [NVIDIA A100](https://www.nvidia.com/en-us/data-center/a100/) GPUs for handling the matrix math also called tensor operations. TF32 running on Tensor Cores in A100 GPUs can provide up to 10x speedups compared to single-precision floating-point math (FP32) on Volta GPUs. 
  
 TF32 Tensor Cores can speed up networks using FP32, typically with no loss of accuracy. It is more robust than FP16 for models which require high dynamic range for weights or activations.
  
-For more information, refer to the [TensorFloat-32 in the A100 GPU Accelerates AI Training, HPC up to 20x](#https://blogs.nvidia.com/blog/2020/05/14/tensorfloat-32-precision-format/) blog post.
+For more information, refer to the [TensorFloat-32 in the A100 GPU Accelerates AI Training, HPC up to 20x](https://blogs.nvidia.com/blog/2020/05/14/tensorfloat-32-precision-format/) blog post.
  
 TF32 is supported in the NVIDIA Ampere GPU architecture and is enabled by default.
  

TensorFlow2/Segmentation/UNet_Medical/README.md

@@ -137,11 +137,11 @@ This implementation exploits the TensorFlow Automatic Mixed Precision feature. T
  
 #### Enabling TF32
 
-TensorFloat-32 (TF32) is the new math mode in [NVIDIA A100](#https://www.nvidia.com/en-us/data-center/a100/) GPUs for handling the matrix math also called tensor operations. TF32 running on Tensor Cores in A100 GPUs can provide up to 10x speedups compared to single-precision floating-point math (FP32) on Volta GPUs. 
+TensorFloat-32 (TF32) is the new math mode in [NVIDIA A100](https://www.nvidia.com/en-us/data-center/a100/) GPUs for handling the matrix math also called tensor operations. TF32 running on Tensor Cores in A100 GPUs can provide up to 10x speedups compared to single-precision floating-point math (FP32) on Volta GPUs. 
 
 TF32 Tensor Cores can speed up networks using FP32, typically with no loss of accuracy. It is more robust than FP16 for models which require high dynamic range for weights or activations.
 
-For more information, refer to the [TensorFloat-32 in the A100 GPU Accelerates AI Training, HPC up to 20x](#https://blogs.nvidia.com/blog/2020/05/14/tensorfloat-32-precision-format/) blog post.
+For more information, refer to the [TensorFloat-32 in the A100 GPU Accelerates AI Training, HPC up to 20x](https://blogs.nvidia.com/blog/2020/05/14/tensorfloat-32-precision-format/) blog post.
 
 TF32 is supported in the NVIDIA Ampere GPU architecture and is enabled by default.