Using all channels by default.

pes_to_spec/model.py

@@ -8,6 +8,7 @@ from scipy.optimize import fmin_l_bfgs_b
 from sklearn.decomposition import PCA, IncrementalPCA
 from sklearn.model_selection import train_test_split
 from sklearn.base import TransformerMixin, BaseEstimator
+from itertools import product
 
 import matplotlib.pyplot as plt
 
@@ -124,17 +125,13 @@ class Model(TransformerMixin, BaseEstimator):
 
     """
     def __init__(self,
-                 channels:List[str]=["channel_1_D",
-                                     "channel_2_B",
-                                     "channel_3_A",
-                                     "channel_3_B",
-                                     "channel_4_C",
-                                     "channel_4_D"],
-                 n_pca_lr: int=400,
+                 channels:List[str]=[f"channel_{j}_{k}"
+                                     for j, k in product(range(1, 5), ["A", "B", "C", "D"])],
+                 n_pca_lr: int=600,
                  n_pca_hr: int=20,
                  high_res_sigma: float=0.2,
-                 tof_start: Optional[int]=31445,
-                 delta_tof: Optional[int]=200,
+                 tof_start: Optional[int]=None,
+                 delta_tof: Optional[int]=300,
                  validation_size: float=0.05):
         self.channels = channels
         self.n_pca_lr = n_pca_lr

pes_to_spec/test/offline_analysis.py

@@ -109,17 +109,19 @@ def main():
     #retvol_raw = run["SA3_XTD10_PES/MDL/DAQ_MPOD", "u212.value"].select_trains(by_id[tids]).ndarray()
     #retvol_raw_timestamp = run["SA3_XTD10_PES/MDL/DAQ_MPOD", "u212.timestamp"].select_trains(by_id[tids]).ndarray()
 
-    model = Model(channels=["channel_1_D",
-                            "channel_2_B",
-                            "channel_3_A",
-                            "channel_3_B",
-                            "channel_4_C",
-                            "channel_4_D"],
+    # these have been manually selected:
+    #useful_channels = ["channel_1_D",
+    #                  "channel_2_B",
+    #                  "channel_3_A",
+    #                  "channel_3_B",
+    #                  "channel_4_C",
+    #                  "channel_4_D"]
+    model = Model(channels=channels,
                  n_pca_lr=400,
                  n_pca_hr=20,
                  high_res_sigma=0.2,
                  tof_start=None,
-                 delta_tof=200,
+                 delta_tof=300,
                  validation_size=0.05)
 
     train_idx = np.isin(tids, train_tids)