2024-03-02T23-22-48Z

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+data/kelly.db filter=lfs diff=lfs merge=lfs -text

Dockerfile

@@ -0,0 +1,20 @@
+FROM python:3.11
+
+WORKDIR /code
+
+COPY ./requirements.txt /code/requirements.txt
+
+RUN pip install --no-cache-dir --upgrade -r /code/requirements.txt
+
+# ADD https://static.simonwillison.net/static/2023/history.db /code/history.db
+COPY ./start.sh /code/start.sh
+COPY ./data /code/data
+
+# RUN sqlite-utils tables /code/history.db --counts
+RUN chmod 755 -R /code/data
+RUN chmod 755 /code/start.sh
+
+COPY ./metadata.yml /code/metadata.yml
+
+# CMD ["datasette", "/code/data/*.db", "-m", "/code/metadata.yml", "--host", "0.0.0.0", "--port", "7860"]
+CMD ["bash", "start.sh"]

Makefile

@@ -0,0 +1,17 @@
+REPO_NAME = biases-llm-reference-letters-datasette
+HF_USERNAME = tddschn
+# https://tddschn-biases-llm-reference-letters-datasette.hf.space/
+HF_DOT_SPACE_URL = https://$(HF_USERNAME)-$(REPO_NAME).hf.space
+
+push-hf:
+	git push hf main
+
+commit-and-push-hf:
+	git-pp
+	git push hf main
+
+open-hf-space:
+	open $(HF_DOT_SPACE_URL)
+
+publish-vercel:
+	datasette publish vercel -m metadata.yml data/*.db --project biases-llm-reference-letters-datasette-vercel

data/kelly.db

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2cb072ad178a2db23dc3e883c60fe2076887ed9a785e53cf88f38b80fe083915
+size 859103232

metadata.yml

@@ -0,0 +1,12 @@
+title: '"Kelly is a warm person" paper data (FULL)'
+about: uclanlp/biases-llm-reference-letters
+about_url: https://github.com/uclanlp/biases-llm-reference-letters
+description_html: |-
+    # <p>Shorter URL available: <a href="https://kelly.datasette.teddysc.me/">kelly.datasette.teddysc.me</a></p>
+    # <br/>
+    # <p>The original CSV files are too large, I selected only 100 rows of each file</p>
+    # <p>See also <a href="https://github.com/tddschn/llm-gender-bias-public/blob/master/papers/kelly-is-a-warm-person.md">the notes on GitHub, including link to paper</a>.</p>
+    # <br/>
+    <p>FULL kelly paper dataset generated from all CSV files in the repository</p>
+    <p>This site is published by <a href="https://github.com/tddschn">Teddy</a></p>
+

requirements.txt

@@ -0,0 +1,5 @@
+datasette
+sqlite-utils
+datasette-vega
+datasette-graphql
+datasette-search-all

start.sh

@@ -0,0 +1,5 @@
+#!/usr/bin/env bash
+
+# CMD ["datasette", "/code/data/*.db", "-m", "/code/metadata.yml", "--host", "0.0.0.0", "--port", "7860"]
+
+datasette /code/data/*.db -m /code/metadata.yml --host "0.0.0.0" --port 7860