Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

NumerFrame #9

Merged
merged 7 commits into from
Feb 2, 2022
Merged

NumerFrame #9

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
b1d54f1
1st version of NumerFrame and refactor preprocessing
CarloLepelaars Feb 1, 2022
eaad6e9
Refactor everything for NumerFrame. All tests passing again.
CarloLepelaars Feb 1, 2022
c9d01bd
Clean up old Dataset functionality.
CarloLepelaars Feb 1, 2022
fd3aae4
Fix library references for numerframe
CarloLepelaars Feb 1, 2022
bf098d4
Refactor for numerframe in index.
CarloLepelaars Feb 1, 2022
6e32992
Fix index imports
CarloLepelaars Feb 1, 2022
f6bd1fc
Update docs for NumerFrame setup. Evaluator docs
CarloLepelaars Feb 2, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,7 +15,7 @@ Example and educational notebooks can be found in the `edu_nbs` directory. Devel
The library features the following tools to build your Numerai pipelines:

- Downloaders
- Dataset
- NumerFrame
- Preprocessing
- Model
- ModelPipeline and ModelPipelineCollection
Expand All @@ -37,7 +37,7 @@ The library features the following tools to build your Numerai pipelines:
#
# # Initialize Dataset
# metadata = {"version": 2, "model_name": "MY_MODEL"}
# dataset = create_dataset(file_path="data/current_round/numerai_tournament_data.parquet", **metadata)
# dataset = create_numerframe(file_path="data/current_round/numerai_tournament_data.parquet", **metadata)
#
# # Define and run pipeline
# model1 = JoblibModel(model_directory="dir_with_joblib_models",
Expand Down
4 changes: 2 additions & 2 deletions docs/_data/sidebars/home_sidebar.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,8 +13,8 @@ entries:
title: Download
url: download.html
- output: web,pdf
title: Dataset
url: dataset.html
title: NumerFrame
url: numerframe.html
- output: web,pdf
title: Preprocessing
url: preprocessing.html
Expand Down
100 changes: 69 additions & 31 deletions docs/evaluation.html
View file
Open in desktop

Large diffs are not rendered by default.

24 changes: 12 additions & 12 deletions docs/index.html
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -66,7 +66,7 @@ <h3 id="2.1.-Contents">2.1. Contents<a class="anchor-link" href="#2.1.-Contents"
<div class="text_cell_render border-box-sizing rendered_html">
<ul>
<li>Downloaders</li>
<li>Dataset</li>
<li>NumerFrame</li>
<li>Preprocessing</li>
<li>Model</li>
<li>ModelPipeline and ModelPipelineCollection</li>
Expand Down Expand Up @@ -105,7 +105,7 @@ <h4 id="2.2.1.-Numerai-Classic">2.2.1. Numerai Classic<a class="anchor-link" hre
<span class="c1">#</span>
<span class="c1"># # Initialize Dataset</span>
<span class="c1"># metadata = {&quot;version&quot;: 2, &quot;model_name&quot;: &quot;MY_MODEL&quot;}</span>
<span class="c1"># dataset = create_dataset(file_path=&quot;data/current_round/numerai_tournament_data.parquet&quot;, **metadata)</span>
<span class="c1"># dataset = create_numerframe(file_path=&quot;data/current_round/numerai_tournament_data.parquet&quot;, **metadata)</span>
<span class="c1">#</span>
<span class="c1"># # Define and run pipeline</span>
<span class="c1"># model1 = JoblibModel(model_directory=&quot;dir_with_joblib_models&quot;,</span>
Expand Down Expand Up @@ -143,15 +143,15 @@ <h4 id="2.2.1.-Numerai-Classic">2.2.1. Numerai Classic<a class="anchor-link" hre



<div id="a02fc2e0-feec-42d4-83da-a71909aed9de"></div>
<div id="50637cab-ba97-46cd-8904-b6ecccb1d480"></div>
<div class="output_subarea output_javascript ">
<script type="text/javascript">
var element = $('#a02fc2e0-feec-42d4-83da-a71909aed9de');
var element = $('#50637cab-ba97-46cd-8904-b6ecccb1d480');

setTimeout(function() {
var nbb_cell_id = 3;
var nbb_unformatted_code = "# slow\n# Download version 2 data\n# downloader = NumeraiClassicDownloader(\"data\")\n# downloader.download_inference_data(\"current_round\")\n# \n# # Initialize Dataset\n# metadata = {\"version\": 2, \"model_name\": \"MY_MODEL\"}\n# dataset = create_dataset(file_path=\"data/current_round/numerai_tournament_data.parquet\", **metadata)\n# \n# # Define and run pipeline\n# model1 = JoblibModel(model_directory=\"dir_with_joblib_models\",\n# model_name=\"test_model\")\n# pipeline = ModelPipeline(pipeline_name=dataset.base_model_name,\n# preprocessors=[],\n# models=[model1],\n# postprocessors=[FeatureNeutralizer(proportion=0.5)])\n# dataset = pipeline(dataset)\n# \n# # Submit\n# key = load_key_from_json(\"my_key.json\")\n# submittor = NumeraiClassicSubmittor(directory_path=\"sub_current_round\", key=key)\n# submittor.full_submission(dataf=dataset.dataf,\n# cols=\"prediction_test_model_neutralized_0.5\",\n# file_name=f\"{dataset.model_name}.csv\",\n# model_name=dataset.model_name,\n# version=dataset.version\n# )\n# \n# # Remove data and subs\n# downloader.remove_base_directory()\n# submittor.remove_base_directory()";
var nbb_formatted_code = "# slow\n# Download version 2 data\n# downloader = NumeraiClassicDownloader(\"data\")\n# downloader.download_inference_data(\"current_round\")\n#\n# # Initialize Dataset\n# metadata = {\"version\": 2, \"model_name\": \"MY_MODEL\"}\n# dataset = create_dataset(file_path=\"data/current_round/numerai_tournament_data.parquet\", **metadata)\n#\n# # Define and run pipeline\n# model1 = JoblibModel(model_directory=\"dir_with_joblib_models\",\n# model_name=\"test_model\")\n# pipeline = ModelPipeline(pipeline_name=dataset.base_model_name,\n# preprocessors=[],\n# models=[model1],\n# postprocessors=[FeatureNeutralizer(proportion=0.5)])\n# dataset = pipeline(dataset)\n#\n# # Submit\n# key = load_key_from_json(\"my_key.json\")\n# submittor = NumeraiClassicSubmittor(directory_path=\"sub_current_round\", key=key)\n# submittor.full_submission(dataf=dataset.dataf,\n# cols=\"prediction_test_model_neutralized_0.5\",\n# file_name=f\"{dataset.model_name}.csv\",\n# model_name=dataset.model_name,\n# version=dataset.version\n# )\n#\n# # Remove data and subs\n# downloader.remove_base_directory()\n# submittor.remove_base_directory()";
var nbb_unformatted_code = "# slow\n# Download version 2 data\n# downloader = NumeraiClassicDownloader(\"data\")\n# downloader.download_inference_data(\"current_round\")\n#\n# # Initialize Dataset\n# metadata = {\"version\": 2, \"model_name\": \"MY_MODEL\"}\n# dataset = create_numerframe(file_path=\"data/current_round/numerai_tournament_data.parquet\", **metadata)\n#\n# # Define and run pipeline\n# model1 = JoblibModel(model_directory=\"dir_with_joblib_models\",\n# model_name=\"test_model\")\n# pipeline = ModelPipeline(pipeline_name=dataset.base_model_name,\n# preprocessors=[],\n# models=[model1],\n# postprocessors=[FeatureNeutralizer(proportion=0.5)])\n# dataset = pipeline(dataset)\n#\n# # Submit\n# key = load_key_from_json(\"my_key.json\")\n# submittor = NumeraiClassicSubmittor(directory_path=\"sub_current_round\", key=key)\n# submittor.full_submission(dataf=dataset.dataf,\n# cols=\"prediction_test_model_neutralized_0.5\",\n# file_name=f\"{dataset.model_name}.csv\",\n# model_name=dataset.model_name,\n# version=dataset.version\n# )\n#\n# # Remove data and subs\n# downloader.remove_base_directory()\n# submittor.remove_base_directory()";
var nbb_formatted_code = "# slow\n# Download version 2 data\n# downloader = NumeraiClassicDownloader(\"data\")\n# downloader.download_inference_data(\"current_round\")\n#\n# # Initialize Dataset\n# metadata = {\"version\": 2, \"model_name\": \"MY_MODEL\"}\n# dataset = create_numerframe(file_path=\"data/current_round/numerai_tournament_data.parquet\", **metadata)\n#\n# # Define and run pipeline\n# model1 = JoblibModel(model_directory=\"dir_with_joblib_models\",\n# model_name=\"test_model\")\n# pipeline = ModelPipeline(pipeline_name=dataset.base_model_name,\n# preprocessors=[],\n# models=[model1],\n# postprocessors=[FeatureNeutralizer(proportion=0.5)])\n# dataset = pipeline(dataset)\n#\n# # Submit\n# key = load_key_from_json(\"my_key.json\")\n# submittor = NumeraiClassicSubmittor(directory_path=\"sub_current_round\", key=key)\n# submittor.full_submission(dataf=dataset.dataf,\n# cols=\"prediction_test_model_neutralized_0.5\",\n# file_name=f\"{dataset.model_name}.csv\",\n# model_name=dataset.model_name,\n# version=dataset.version\n# )\n#\n# # Remove data and subs\n# downloader.remove_base_directory()\n# submittor.remove_base_directory()";
var nbb_cells = Jupyter.notebook.get_cells();
for (var i = 0; i < nbb_cells.length; ++i) {
if (nbb_cells[i].input_prompt_number == nbb_cell_id) {
Expand Down Expand Up @@ -227,10 +227,10 @@ <h4 id="2.2.1.-Numerai-Classic">2.2.1. Numerai Classic<a class="anchor-link" hre



<div id="3588cfe1-c780-4ff5-8d14-8678fd740163"></div>
<div id="c7caa384-7b03-4edb-aa3a-fb38eba90c92"></div>
<div class="output_subarea output_javascript ">
<script type="text/javascript">
var element = $('#3588cfe1-c780-4ff5-8d14-8678fd740163');
var element = $('#c7caa384-7b03-4edb-aa3a-fb38eba90c92');

setTimeout(function() {
var nbb_cell_id = 4;
Expand Down Expand Up @@ -286,10 +286,10 @@ <h3 id="2.2.2.-Numerai-Signals">2.2.2. Numerai Signals<a class="anchor-link" hre



<div id="0d720b41-d343-4395-9d96-16ed8f396180"></div>
<div id="31b7b1a6-04c4-4407-9252-2ad3242e5588"></div>
<div class="output_subarea output_javascript ">
<script type="text/javascript">
var element = $('#0d720b41-d343-4395-9d96-16ed8f396180');
var element = $('#31b7b1a6-04c4-4407-9252-2ad3242e5588');

setTimeout(function() {
var nbb_cell_id = 5;
Expand Down Expand Up @@ -364,10 +364,10 @@ <h3 id="Branch-structure">Branch structure<a class="anchor-link" href="#Branch-s



<div id="dfb6d8c1-2b5e-40cf-993b-1a1cef536140"></div>
<div id="9764683e-c74d-4766-a193-d93eb44252f2"></div>
<div class="output_subarea output_javascript ">
<script type="text/javascript">
var element = $('#dfb6d8c1-2b5e-40cf-993b-1a1cef536140');
var element = $('#9764683e-c74d-4766-a193-d93eb44252f2');

setTimeout(function() {
var nbb_cell_id = 6;
Expand Down
Loading