change conversion

app.py

@@ -19,7 +19,7 @@ from pytorch_forecasting import (
 from PIL import Image
 
 ## Functions 
-def raw_preds_to_df(raw, idx_offset, quantiles = None):
+def raw_preds_to_df(raw, quantiles = None):
     """
     raw is output of model.predict with return_index=True
     quantiles can be provided like [0.1,0.5,0.9] to get interpretable quantiles
@@ -31,15 +31,13 @@ def raw_preds_to_df(raw, idx_offset, quantiles = None):
     dec_len = raw.output.prediction.shape[1]
     n_quantiles = preds.shape[-1]
     preds_df = pd.DataFrame(index.values.repeat(dec_len * n_quantiles, axis=0),columns=index.columns)
-    preds_df = preds_df.assign(Horizon=np.tile(np.repeat(np.arange(1,1+dec_len),n_quantiles),len(preds_df)//(dec_len*n_quantiles)))
-    preds_df = preds_df.assign(Quantile=np.tile(np.arange(n_quantiles),len(preds_df)//n_quantiles))
-    preds_df = preds_df.assign(Prediction=preds.flatten().cpu().numpy())
+    preds_df = preds_df.assign(h=np.tile(np.repeat(np.arange(1,1+dec_len),n_quantiles),len(preds_df)//(dec_len*n_quantiles)))
+    preds_df = preds_df.assign(q=np.tile(np.arange(n_quantiles),len(preds_df)//n_quantiles))
+    preds_df = preds_df.assign(pred=preds.flatten().cpu().numpy())
     if quantiles is not None:
-        preds_df['Quantile'] = preds_df['Quantile'].map({i:q for i,q in enumerate(quantiles)})
+        preds_df['q'] = preds_df['q'].map({i:q for i,q in enumerate(quantiles)})
 
-    preds_df['pred_idx'] = preds_df['time_idx'] + preds_df['Horizon'] - 1
-    preds_df['Date'] = pd.to_datetime(idx_offset)
-    preds_df['Date'] = preds_df['Date'] + preds_df['pred_idx'].apply(pd.DateOffset)
+    preds_df['pred_idx'] = preds_df['time_idx'] + preds_df['h'] - 1
     return preds_df
 
 def prepare_dataset(_parameters, df, rain, temperature, datepicker, mapping):
@@ -57,10 +55,9 @@ def prepare_dataset(_parameters, df, rain, temperature, datepicker, mapping):
     df = TimeSeriesDataSet.from_parameters(_parameters, df)
     return df.to_dataloader(train=False, batch_size=256,num_workers = 0)
 
-def predict(_model, _dataloader, datepicker, df):
-    out = _model.predict(_dataloader, mode="raw", return_index=True)# return_x=True, 
-    first_date = min(df["Date"])
-    preds = raw_preds_to_df(raw = out, idx_offset = first_date)
+def predict(_model, _dataloader):
+    out = model.predict(_dataloader, mode="raw", return_x=True,return_index=True, trainer_kwargs=dict(accelerator="cpu"))
+    preds = raw_preds_to_df(out)
     return preds[["pred_idx", "Group", "pred"]]
 
 def adjust_data_for_plot(df, preds):
@@ -157,7 +154,7 @@ def main():
     rain = st.selectbox("Rain Indicator", ('Default', 'Yes', 'No'), key = "rain")
 
     dataloader = prepare_dataset(parameters, df.copy(), st.session_state.rain, st.session_state.temperature, st.session_state.date, RAIN_MAPPING)
-    preds = predict(model, dataloader, st.session_state.date, df)
+    preds = predict(model, dataloader)
 
     data_plot = adjust_data_for_plot(df.copy(), preds)
     fig, _ = generate_plot(data_plot)