update space

.gitignore

@@ -7,4 +7,6 @@ files/
 .mamba-env/
 .mamba-env-cache/
 .DS_Store
-.vscode/
+.vscode/
+.gradio/
+__pycache__

app.py

@@ -3,26 +3,61 @@ from tools import *
 from format_dexcom import process_csv
 import tempfile
 import os
+from huggingface_hub import list_models
+from typing import List, Tuple
+from pathlib import Path
+import plotly.graph_objects as go
+from huggingface_hub import HfApi
 
-def process_and_prepare(file):
-    """Process the raw CSV and prepare it for prediction"""
-    # Create a temporary file for the processed CSV
+def get_available_models() -> List[str]:
+    """Get list of available gluformer models from HuggingFace."""
+    api = HfApi()
+    files = api.list_repo_files("Livia-Zaharia/gluformer_models")
+    
+    # Filter for .pth files
+    gluformer_models = [
+        file for file in files 
+        if file.endswith('.pth') and "weights" in file.lower() and 'gluformer' in file.lower()
+    ]
+    
+    return gluformer_models
+
+AVAILABLE_MODELS = get_available_models()
+print(AVAILABLE_MODELS)
+
+def process_and_prepare(file: tempfile._TemporaryFileWrapper, model_name: str) -> Tuple[gr.Slider, gr.Markdown]:
+    """Process the raw CSV and prepare it for prediction.
+    
+    Args:
+        file: Uploaded temporary file object
+        model_name: Name of the selected model
+        
+    Returns:
+        Tuple containing:
+        - Updated slider component
+        - Sample count markdown component
+    """
     with tempfile.NamedTemporaryFile(delete=False, suffix='.csv') as tmp_file:
         processed_path = tmp_file.name
     
-    # Process the CSV file
     process_csv(
         input_dir=file.name,
         output_file=processed_path
     )
     
-    # Run the preparation step with processed file
-    return prep_predict_glucose_tool(processed_path)
+    return prep_predict_glucose_tool(processed_path, model_name)
+
 
 with gr.Blocks() as demo:
     gr.Markdown("# Glucose Prediction Tool")
     gr.Markdown("Upload a Dexcom CSV file to get predictions")
     
+    model_selector = gr.Dropdown(
+        choices=AVAILABLE_MODELS,
+        value="gluformer_1samples_500epochs_10heads_32batch_geluactivation_livia_large_weights.pth",
+        label="Select Model",
+        interactive=True
+    )
     file_input = gr.File(label="Upload Raw Dexcom CSV File")
     with gr.Row():
         index_slider = gr.Slider(
@@ -39,18 +74,11 @@ with gr.Blocks() as demo:
     # Update slider and show total samples when file is uploaded
     file_input.change(
         fn=process_and_prepare,
-        inputs=[file_input],
+        inputs=[file_input, model_selector],
         outputs=[index_slider, sample_count],
         queue=True
     )
     
-    # Update visibility after processing
-    file_input.change(
-        fn=lambda: (gr.Slider(visible=True), gr.Markdown(visible=True)),
-        outputs=[index_slider, sample_count],
-        queue=True
-    )
-
     # Only update plot after processing is complete
     index_slider.change(
         fn=predict_glucose_tool,
@@ -59,4 +87,4 @@ with gr.Blocks() as demo:
         queue=True
     )
 
-demo.launch()
+demo.launch(share=True)

requirements.txt

@@ -13,4 +13,5 @@ transformers
 datasets
 python-multipart
 plotly
-kaleido
+kaleido
+huggingface-hub

test_environment.yaml

@@ -0,0 +1,23 @@
+name: glucose
+channels:
+  - conda-forge
+  - defaults
+dependencies:
+  - python=3.11
+  - pip:
+    - torch
+    - optuna
+    - numpy==1.26.4
+    - tensorboard
+    - pandas
+    - psutil
+    - typer
+    - darts==0.29.0
+    - pmdarima==2.0.4
+    - peft
+    - transformers
+    - datasets
+    - python-multipart
+    - plotly
+    - kaleido
+    - huggingface-hub

tools.py

@@ -15,12 +15,16 @@ from huggingface_hub import hf_hub_download
 import plotly.graph_objects as go
 import gradio as gr
 from format_dexcom import *
+from typing import Tuple, Union, List
+from plotly.graph_objs._figure import Figure
+from gradio.components import Slider
+from gradio.components import Markdown
 
 
 glucose = Path(os.path.abspath(__file__)).parent.resolve()
 file_directory = glucose / "files"
        
-def plot_forecast(forecasts: np.ndarray, filename: str,ind:int=10):
+def plot_forecast(forecasts: np.ndarray, filename: str,ind:int=10) -> Tuple[Path, Figure]:
     
     forecasts = (forecasts - scalers['target'].min_) / scalers['target'].scale_
 
@@ -134,14 +138,14 @@ scalers = None
 dataset_test_glufo = None
 filename = None
 
-def prep_predict_glucose_tool(file):
+def prep_predict_glucose_tool(file: Union[str, Path], model_name: str = "gluformer_1samples_10000epochs_10heads_32batch_geluactivation_livia_mini_weights.pth") -> Tuple[Slider, Markdown]:
     """
     Function to predict future glucose of user.
     """
     global formatter, series, scalers, glufo, dataset_test_glufo, filename
     
     model = "Livia-Zaharia/gluformer_models"
-    model_path = hf_hub_download(repo_id=model, filename="gluformer_1samples_10000epochs_10heads_32batch_geluactivation_livia_mini_weights.pth")
+    model_path = hf_hub_download(repo_id=model, filename=model_name)
     
     formatter, series, scalers = load_data(
         url=str(file), 
@@ -198,20 +202,23 @@ def prep_predict_glucose_tool(file):
     filename = generate_filename_from_url(file)
 
     max_index = len(dataset_test_glufo) - 1
+
+    print(f"Total number of test samples: {max_index + 1}")
     
     return (
         gr.Slider(
             minimum=0,
-            maximum=max_index,
-            value=10,
+            maximum=max_index-1,
+            value=max_index,
             step=1,
             label="Select Sample Index",
+            visible=True
         ),
-        gr.Markdown(f"Total number of test samples: {max_index + 1}")
+        gr.Markdown(f"Total number of test samples: {max_index + 1}", visible=True)
     )
 
 
-def predict_glucose_tool(ind) -> go.Figure:
+def predict_glucose_tool(ind: int) -> Figure:
        
     
     device = "cuda" if torch.cuda.is_available() else "cpu"