Merge branch 'master' of https://github.com/biswajitsahoo1111/cbm_cod…

…es_open

Just merge it.

_includes/utterances.html

@@ -0,0 +1,8 @@
+<script
+  async
+  src="https://utteranc.es/client.js"
+  repo="biswajitsahoo1111/cbm_codes_open"
+  issue-term="title"
+  theme="github-light"
+  crossorigin="anonymous"
+></script>

_layouts/default.html

@@ -47,6 +47,8 @@ <h2 class="project-tagline">{{ site.description | default: site.description | de
 
     <main id="content" class="main-content" role="main">
       {{ content }}
+
+      {% include utterances.html %}
 
       <footer class="site-footer">
          <!--

index.md

@@ -17,7 +17,7 @@ This is an ongoing project and modifications and additions of new techniques wil
 
 ## Results using [Case Western Reserve University Bearing Data](https://csegroups.case.edu/bearingdatacenter/pages/welcome-case-western-reserve-university-bearing-data-center-website)<sup>*</sup>
 -------------------------------
-We will first apply classical feature based methods (so-called shallow learning methods) to obtain results and then apply deep learning based methods. In feature based methods, we will extensively use [wavelet packet energy features](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/calculate_wavelet_packet_energy_features.ipynb) and [wavelet packet entropy featues](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/calculate_wavelet_packet_entropy_features.ipynb) that are calculated from raw time domain data. Dataset description and [time domain preprocessing](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/CWRU_time_domain_data_preprocessing.ipynb) steps can be found [here](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/CWRU_time_domain_data_preprocessing.ipynb). Steps to compute time domain features are explained in [this notebook](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/Calculating_time_domain_features_CWRU.ipynb). The procedure detailing calculation of wavelet packet energy features can be found at [this link](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/calculate_wavelet_packet_energy_features.ipynb) and similar calculations for wavelet packet entropy features can be found at [this link](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/calculate_wavelet_packet_entropy_features.ipynb). Also see the following two notebooks for computation of wavelet packet features in Python: [Wavelet packet energy features in Python](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/Wavelet_packet_energy_features_python.ipynb) and [Wavelet packet entropy features in Python](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/Wavelet_packet_entropy_features_python.ipynb).
+We will first apply classical feature based methods (so-called shallow learning methods) to obtain results and then apply deep learning based methods. In feature based methods, we will extensively use [wavelet packet energy features](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/calculate_wavelet_packet_energy_features.ipynb) and [wavelet packet entropy featues](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/calculate_wavelet_packet_entropy_features.ipynb) that are calculated from raw time domain data. Dataset description and [time domain preprocessing](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/CWRU_time_domain_data_preprocessing.ipynb) steps can be found [here](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/CWRU_time_domain_data_preprocessing.ipynb). Steps to [compute time domain features](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/Calculating_time_domain_features_CWRU.ipynb) are explained in [this notebook](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/Calculating_time_domain_features_CWRU.ipynb). The procedure detailing calculation of wavelet packet energy features can be found at [this link](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/calculate_wavelet_packet_energy_features.ipynb) and similar calculations for wavelet packet entropy features can be found at [this link](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/calculate_wavelet_packet_entropy_features.ipynb). Also see the following two notebooks for computation of wavelet packet features in Python: [Wavelet packet energy features in Python](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/Wavelet_packet_energy_features_python.ipynb) and [Wavelet packet entropy features in Python](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/Wavelet_packet_entropy_features_python.ipynb).
 
 1. [SVM on time domain
     features](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/SVM_multiclass_time_cwru_python.ipynb) (10 classes, sampling frequency: 48k) (Overall accuracy: **96.5%**) ([Python code](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/SVM_multiclass_time_cwru_python.ipynb)) ([R code](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/SVM_multiclass_time.pdf))
@@ -82,17 +82,19 @@ In this section, we will show results of fault diagnosis task using deep learnin
 
 -------------------------------
 
-1. [Data-Driven Remaining Useful Life (RUL) Prediction](https://biswajitsahoo1111.github.io/rul_codes_open/) 
+1. [Tensorflow 2 code for Attention Mechanisms chapter of Dive into Deep Learning (D2L) book.](https://github.com/biswajitsahoo1111/D2L_Attention_Mechanisms_in_TF)
 
-2. [Fault diagnosis of machines (A non-technical introduction)](https://biswajitsahoo1111.github.io/post/fault-diagnosis-of-machines/)
+2. [Data-Driven Remaining Useful Life (RUL) Prediction](https://biswajitsahoo1111.github.io/rul_codes_open/) 
 
-3. [Blog articles by yours truly](https://biswajitsahoo1111.github.io/categories/blog/)
+3. [Fault diagnosis of machines (A non-technical introduction)](https://biswajitsahoo1111.github.io/post/fault-diagnosis-of-machines/)
 
-4. [A quick introduction to MATLAB](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/matlab_intro.pdf)
+4. [Blog articles by yours truly](https://biswajitsahoo1111.github.io/categories/blog/)
 
-5. [Transient vibration and shock response spectrum plots in MATLAB](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/transient_vibration_and_SRS_plots.pdf)
+5. [A quick introduction to MATLAB](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/matlab_intro.pdf)
 
-6. [Simple examples on finding instantaneous frequency using Hilbert transform](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/hilbert_inst_freq_modulation.pdf) ([MATLAB Code](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/hilbert_inst_freq_modulation.pdf))
+6. [Transient vibration and shock response spectrum plots in MATLAB](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/transient_vibration_and_SRS_plots.pdf)
+
+7. [Simple examples on finding instantaneous frequency using Hilbert transform](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/hilbert_inst_freq_modulation.pdf) ([MATLAB Code](https://github.com/biswajitsahoo1111/cbm_codes_open/blob/master/notebooks/hilbert_inst_freq_modulation.pdf))
 
 ------------------------------
 

notebooks/Reading_multiple_files_in_Tensorflow_2.ipynb

@@ -1309,17 +1309,20 @@
     "\n",
     "I hope, this blog will be of help to readers. Please bring any errors or omissions to my notice.\n",
     "\n",
-    "**Update**: If along with reading, one has to perform complex transformations on extracted data (say, doing spectrogram on each segment of data, etc.), the naive approach presented in this blog may turn out to be slow. But there are ways to make these computations faster. One such speedup technique can be found at [this blog](https://biswajitsahoo1111.github.io/post/efficiently-reading-multiple-files-in-tensorflow-2/).\n",
+    "**Update 1**: While generators are convenient for handling chunks of data from a large dataset, they have limited portability and scalability. Therefore, in Tensorflow Sequences are preferred instead of generators. [See this blog](https://biswajitsahoo1111.github.io/post/reading-multiple-files-in-tensorflow-2-using-sequence/) for a complete workflow for reading multiple files using Sequence.\n",
+    "\n",
+    "\n",
+    "**Update 2**: If along with reading, one has to perform complex transformations on extracted data (say, doing spectrogram on each segment of data, etc.), the naive approach presented in this blog may turn out to be slow. But there are ways to make these computations faster. One such speedup technique can be found at [this blog](https://biswajitsahoo1111.github.io/post/efficiently-reading-multiple-files-in-tensorflow-2/). Readers should also try using [Sequence](https://biswajitsahoo1111.github.io/post/reading-multiple-files-in-tensorflow-2-using-sequence/) and see if it improves input pipeline efficiency.\n",
     "\n",
     "Last modified: 27th April, 2020."
    ]
   }
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "tf_cpu_22",
+   "display_name": "tf_24_env_cpu",
    "language": "python",
-   "name": "tf_cpu_22"
+   "name": "tf_24_env_cpu"
   },
   "language_info": {
    "codemirror_mode": {
@@ -1331,7 +1334,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.7"
+   "version": "3.8.5"
   },
   "toc": {
    "base_numbering": 1,