add figs and deploy

Author: chiral-carbon

.github/workflows/update_space.yml

@@ -1,28 +1,82 @@
-name: Run Python script
+name: Deploy to Hugging Face Spaces
 
 on:
   push:
     branches:
       - main
 
 jobs:
-  build:
+  deploy:
     runs-on: ubuntu-latest
 
     steps:
     - name: Checkout
-      uses: actions/checkout@v2
+      uses: actions/checkout@v3
 
-    - name: Set up Python
-      uses: actions/setup-python@v2
+    - name: Set up Conda
+      uses: conda-incubator/setup-miniconda@v2
       with:
-        python-version: '3.9'
+        activate-environment: kg4s
+        environment-file: environment.yml
+        auto-activate-base: false
+        use-mamba: true
 
-    - name: Install Gradio
-      run: python -m pip install gradio
+    - name: Verify Conda installation
+      shell: bash -l {0}
+      run: |
+        conda info
+        conda list
 
-    - name: Log in to Hugging Face
-      run: python -c 'import huggingface_hub; huggingface_hub.login(token="${{ secrets.hf_token }}")'
+    - name: Install huggingface_hub
+      shell: bash -l {0}
+      run: |
+        pip install huggingface_hub
+        pip list
 
-    - name: Deploy to Spaces
-      run: gradio deploy
+    - name: Deploy to Hugging Face Spaces
+      env:
+        HF_TOKEN: ${{ secrets.HF_TOKEN }}
+      shell: bash -l {0}
+      run: |
+        python - <<EOF
+        import os
+        import sys
+        from huggingface_hub import HfApi
+        
+        print("Python script started")
+        print(f"Python version: {sys.version}")
+        print(f"Current working directory: {os.getcwd()}")
+        print(f"Contents of current directory: {os.listdir('.')}")
+        
+        sys.path.append('scripts')
+        print(f"Updated sys.path: {sys.path}")
+        print(f"Contents of scripts directory: {os.listdir('scripts')}")
+        
+        print("Importing demo from run_db_interface")
+        from run_db_interface import demo
+        print("Demo imported successfully")
+        
+        api = HfApi()
+        print("HfApi initialized")
+        
+        print("Creating/verifying repository")
+        api.create_repo(
+            repo_id="abby101/xurveyor-0",
+            repo_type="space",
+            space_sdk="gradio",
+            token="$HF_TOKEN"
+        )
+        print("Repository created or verified")
+        
+        print("Starting deployment")
+        demo.deploy(
+            repo_id="abby101/xurveyor-0",
+            hf_token="$HF_TOKEN",
+        )
+        print("Deployment completed")
+        EOF
+
+    - name: Check Hugging Face Space
+      run: |
+        echo "Deployment process completed. Please check your Hugging Face Space at https://huggingface.co/spaces/abby101/xurveyor-0"
+        echo "If the space is not updated, please check the logs above for any errors."

.gitignore

@@ -165,5 +165,7 @@ requirements.txt
 wandb/
 slurm_logs/
 notebooks/
-misc/
+misc/polymathic_data_files
+misc/notes
+misc/test.ipynb
 Meta-Llama-3-70B-Instruct/

README.md

@@ -1,9 +1,3 @@
----
-title: git_config_-global_credential.helper_store
-app_file: scripts/run_db_interface.py
-sdk: gradio
-sdk_version: 4.40.0
----
 # Mapping the Data Landscape For Generalizable Scientific Models
 
 This is a WIP that builds a knowledge base to store structured information extracted from scientific publications, datasets and articles using LLMs. 
@@ -14,25 +8,29 @@ This tool helps us identify the gaps where current foundation models lack covera
 
 We use the Llama-3-70B-Instruct model for structured information extraction. 
 
-<div style="display: flex; justify-content: space-between; gap: 20px;">
-    <figure style="margin: 0; width: 48%;">
-        <img src="misc/eval_pipeline.png" alt="Fig 1" style="width: 100%; height: 300px; object-fit: contain;">
-        <figcaption style="font-size: 0.9em; text-align: center; margin-top: 10px;">
-            Prompt optimization pipeline to maximize precision of the model annotated 
-            predictions by running on manually annotated subset of scientific corpora. 
-            The tagged outputs can be generated as JSON or in a readable format, and be 
-            generated using temperature and nucleus sampling (sweep hyperparams).
-        </figcaption>
-    </figure>
-    <figure style="margin: 0; width: 48%;">
-        <img src="misc/pipeline.png" alt="Fig 2" style="width: 100%; height: 300px; object-fit: contain;">
-        <figcaption style="font-size: 0.9em; text-align: center; margin-top: 10px;">
-            Illustration of the structured prediction pipeline on the full corpus of 
-            scientific papers, which runs optimized prompts and stores the model's 
-            outputs in a SQL db.
-        </figcaption>
-    </figure>
-</div>
+## Workflow
+
+<table>
+  <tr>
+    <td width="50%" valign="top">
+      <img src="misc/eval_pipeline.png" alt="Fig 1" width="100%">
+      <p align="center">
+        <em>Fig 1: Prompt optimization pipeline to maximize precision of the model annotated 
+        predictions by running on manually annotated subset of scientific corpora. The 
+        tagged outputs can be generated as JSON or in a readable format, and be 
+        generated using temperature and nucleus sampling (sweep hyperparams).</em>
+      </p>
+    </td>
+    <td width="50%" valign="top">
+      <img src="misc/pipeline.png" alt="Fig 2" width="100%">
+      <p align="center">
+        <em>Fig 2: Illustration of the structured prediction pipeline on the full corpus of 
+        scientific papers, which runs optimized prompts and stores the model's outputs in 
+        a SQL db.</em>
+      </p>
+    </td>
+  </tr>
+</table>
 
 ## Installation