stooq data and other updates

Author: stefan-jansen

21_gans_for_synthetic_time_series/02_TimeGAN_TF2.ipynb

@@ -1083,9 +1083,9 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python [conda env:ml4t-dl]",
+   "display_name": "Python 3",
    "language": "python",
-   "name": "conda-env-ml4t-dl-py"
+   "name": "python3"
   },
   "language_info": {
    "codemirror_mode": {
@@ -1097,7 +1097,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.7"
+   "version": "3.8.5"
   },
   "toc": {
    "base_numbering": 1,

21_gans_for_synthetic_time_series/03_evaluating_synthetic_data.ipynb

@@ -20,6 +20,7 @@ A few sample data sources that we will source and work with include, among other
 There are several notebooks that guide you through the data sourcing process:
 - The notebook [create_datasets](create_datasets.ipynb) contains information on downloading the **Quandl Wiki stock prices** and a few other sources that we use throughout the book, such as S&P500 benchmark, and US equities metadata.
 - The notebook [create_stooq_data](create_stooq_data.ipynb) demonstrates how to download historical prices for Japanese stocks and US stocks and ETFs from STOOQ.
+  > Please note that STOOQ will disable automatic downloads and require CAPTCHA starting Dec 10, 2020 so that the code that downloads and unpacks the zip files will no longer work; please navigate to their website for manual download.
 - The notebook [create_yelp_review_data](create_yelp_review_data.ipynb) combines text data with additional numerical features for sentiment analysis from Yelp user reviews. 
 - The notebook [glove_word_vectors](glove_word_vectors.ipynb) downloads pre-trained word vectors.
 - The notebook [twitter_sentiment](twitter_sentiment.ipynb) downloads and extracts twitter data for sentiment analysis.

data/README.md

@@ -99,7 +99,9 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "> Note that the below downloading details may change at any time as Stooq updates their website; if you encounter errors, please inspect their website and raise a GitHub issue to let us know so we can update the information."
+    "> Note that the below downloading details may change at any time as Stooq updates their website; if you encounter errors, please inspect their website and raise a GitHub issue to let us know so we can update the information.\n",
+    "\n",
+    "> Update 12/2020: please note that STOOQ will disable automatic downloads and require CAPTCHA starting Dec 10, 2020 so that the code that downloads and unpacks the zip files will no longer work; please navigate to their website using the below link for manual download. "
    ]
   },
   {
@@ -1338,4 +1340,4 @@
  },
  "nbformat": 4,
  "nbformat_minor": 4
-}
+}

data/create_stooq_data.ipynb

@@ -143,7 +143,7 @@ Now you are running a shell inside the container and can access the various [con
 | 4         | 17-22*           | ml4t-dl  |
 | all | throughout              | ml4t-zipline
 
-> *the Deep Reinforcement Learning examples require TensorFlow 2.2, which currently is only available via `conda` for GPUC; the notebooks contain instructions for upgrading via `pip`. 
+> the Deep Reinforcement Learning examples require TensorFlow 2.2, which currently is only available for Linux via `conda` for GPU; the notebooks contain instructions for upgrading via `pip`. Check [here](https://anaconda.org/anaconda/tensorflow) for current CPU and [here](https://anaconda.org/anaconda/tensorflow-gpu) for current GPU version support.
 
 - You can switch to another environment using `conda activate <env_name>`.
 - Alternatively, you can switch from one environment to another from the jupyter notebook or jupyter lab thanks to the [nb_conda_kernels](https://github.com/Anaconda-Platform/nb_conda_kernels) extension (see below).

installation/README.md

@@ -245,7 +245,6 @@ dependencies:
   - ta-lib-base=0.4.0
   - tensorboard=2.2.2
   - tensorboard-plugin-wit=1.6.0
-  - tensorflow=2.2.0
   - tensorflow-base=2.2.0
   - tensorflow-datasets=1.2.0
   - tensorflow-estimator=2.2.0