diff --git a/src/methods/lioretn/README.txt b/src/methods/lioretn/README.txt
new file mode 100644
index 0000000000000000000000000000000000000000..adde0a780d46e5cecc6e7622a1d2eb2c521bf4cf
--- /dev/null
+++ b/src/methods/lioretn/README.txt
@@ -0,0 +1 @@
+Quite simple to use : reconstruct.py file calls demosaicking.py (where is made the demosaicking) and is called by the main.ipynb
\ No newline at end of file
diff --git a/src/methods/lioretn/demosaicking.py b/src/methods/lioretn/demosaicking.py
index 7b2ed87c4059aa36a5366c366a51fba9fc8e83f5..1b697d75853c3827ca61f7d5b282cd758843d404 100644
--- a/src/methods/lioretn/demosaicking.py
+++ b/src/methods/lioretn/demosaicking.py
@@ -84,11 +84,13 @@ def high_quality_linear_interpolation(op : CFA, y : np.ndarray) -> np.ndarray:
                 if mask[i, j, G] == 1: # We must estimate R and B components
                     res[i, j, G] = y[i, j]
 
+                    # Estimate R (B at left and right or B at top and bottom)
                     if mask[i, max(0, j-1), R] == 1 or mask[i, min(y.shape[1]-1, j+1), R] == 1:
                         res[i, j, R] = float(convolve2d(y_pad[i_pad-2:i_pad+3, j_pad-2:j_pad+3], bayer_r_at_green_rrow_bcol, mode='valid'))
                     else:
                         res[i, j, R] = float(convolve2d(y_pad[i_pad-2:i_pad+3, j_pad-2:j_pad+3], bayer_r_at_green_brow_rcol, mode='valid'))
 
+                    # Estimate B (R at left and right or R at top and bottom)
                     if mask[i, max(0, j-1), B] == 1 or mask[i, min(y.shape[1]-1, j+1), B] == 1:
                         res[i, j, B] = float(convolve2d(y_pad[i_pad-2:i_pad+3, j_pad-2:j_pad+3], bayer_b_at_green_brow_rcol, mode='valid'))
                     else:
@@ -114,11 +116,13 @@ def high_quality_linear_interpolation(op : CFA, y : np.ndarray) -> np.ndarray:
                 if mask[i, j, G] == 1: # We must estimate R and B components
                     res[i:i+2, j:j+2, G] = y[i:i+2, j:j+2]
 
+                    # Estimate R (B at left and right or B at top and bottom)
                     if mask[i, max(0, j-1), R] == 1 or mask[i, min(y.shape[1]-1, j+1), R] == 1:
                         res[i:i+2, j:j+2, R] = float(convolve2d(y_pad[i_pad-4:i_pad+6, j_pad-4:j_pad+6], quad_r_at_green_rrow_bcol, mode='valid'))
                     else:
                         res[i:i+2, j:j+2, R] = float(convolve2d(y_pad[i_pad-4:i_pad+6, j_pad-4:j_pad+6], quad_r_at_green_brow_rcol, mode='valid'))
 
+                    # Estimate B (R at left and right or R at top and bottom)
                     if mask[i, max(0, j-1), B] == 1 or mask[i, min(y.shape[1]-1, j+1), B] == 1:
                         res[i:i+2, j:j+2, B] = float(convolve2d(y_pad[i_pad-4:i_pad+6, j_pad-4:j_pad+6], quad_b_at_green_brow_rcol, mode='valid'))
                     else:
diff --git a/src/methods/lioretn/reconstruct.py b/src/methods/lioretn/reconstruct.py
index 80f771adc5b319cc3b90196af398a5469dc3e743..5790f4031bd1dbbed43a482c6e827e94d80935ee 100644
--- a/src/methods/lioretn/reconstruct.py
+++ b/src/methods/lioretn/reconstruct.py
@@ -7,7 +7,6 @@ Students can call their functions (declared in others files of src/methods/your_
 import numpy as np
 
 from src.forward_model import CFA
-from src.methods.lioretn.demoisaicing_fct import High_Quality_Linear_Interpolation
 from src.methods.lioretn.demosaicking import high_quality_linear_interpolation
 
 def run_reconstruction(y: np.ndarray, cfa: str) -> np.ndarray:
@@ -26,7 +25,6 @@ def run_reconstruction(y: np.ndarray, cfa: str) -> np.ndarray:
     op = CFA(cfa, input_shape)
 
     res = high_quality_linear_interpolation(op, y)
-    # res = High_Quality_Linear_Interpolation(op, y)
 
     return res