almost finished spectral analysis and creating features from spectral…

… analysis for RNN and CNN

.ipynb_checkpoints/Multifractal Analysis-checkpoint.ipynb

@@ -839,8 +839,7 @@
     "    # Print the head (first 10 elements) of the Hurst exponents array\n",
     "    print(\"Hurst exponents:\", hurst_exponents)\n",
     "\n",
-    "    print(\"Hurst exponents saved to:\", file_path)\n",
-    "\n"
+    "    print(\"Hurst exponents saved to:\", file_path)"
    ]
   },
   {
@@ -986,7 +985,7 @@
     "montage = mne.channels.make_dig_montage(ch_pos=dict(zip(channel_names, scalp_positions)), coord_frame='head')\n",
     "\n",
     "# Set the montage to the Info object\n",
-    "info.set_montage(montage)\n"
+    "info.set_montage(montage)"
    ]
   },
   {
@@ -1067,7 +1066,7 @@
     "# Plot the topographical map of mean Hurst exponent values\n",
     "evoked.plot_topomap(times=0, scalings=1.0, cmap='viridis')\n",
     "\n",
-    "# Consider performing statistical tests (e.g., ANOVA, t-test) to assess group differences or condition-related variations in the mean Hurst exponent values\n"
+    "# Consider performing statistical tests (e.g., ANOVA, t-test) to assess group differences or condition-related variations in the mean Hurst exponent values"
    ]
   },
   {
@@ -1262,6 +1261,14 @@
     "np.save(rnn_save_dir + 'rnn_mfdfa_X_df.npy', rnn_X_data)"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "id": "4469a3d8-c1d4-4f38-b0a2-0ddf202fb728",
+   "metadata": {},
+   "source": [
+    "# Load  Hurst exponents as numpy"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": 18,
@@ -1278,7 +1285,7 @@
    "id": "8f235b5c-b8a9-4a4e-9d28-793fd14d454e",
    "metadata": {},
    "source": [
-    "# Load MFDFA RNN X data"
+    "# Load MFDFA RNN X data as numpy"
    ]
   },
   {
@@ -1295,6 +1302,28 @@
     "rnn_mfdfa_X_df_npy = np.load(rnn_mfdfa_X_df_path + 'rnn_mfdfa_X_df.npy')"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "id": "a6e0d4af-f6ca-468b-8eef-696a7a5a7c8e",
+   "metadata": {},
+   "source": [
+    "# Load MFDFA CNN X data as numpy"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "3411ab35-af06-4d08-87bf-d1e27eacbc68",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Load data\n",
+    "cnn_save_dir = '/home/vincent/AAA_projects/MVCS/Neuroscience/CNN_data/'\n",
+    "\n",
+    "# Load CNN data\n",
+    "rnn_mfdfa_X_df_npy = np.load(rnn_mfdfa_X_df_path + 'cnn_mfdfa_X_df.npy')"
+   ]
+  },
   {
    "cell_type": "markdown",
    "id": "a7f9756f-4857-4c8b-bd70-093c764f9eb5",
@@ -1488,14 +1517,14 @@
    ],
    "source": [
     "# Display the shapes\n",
-    "print(\"Shape of cnn_X_data:\", cnn_X_data.shape)\n",
-    "print(\"Shape of rnn_X_data:\", rnn_X_data.shape)\n",
+    "print(\"Shape of cnn_mfdfa_X_df:\", cnn_mfdfa_X_df.npy.shape)\n",
+    "print(\"Shape of rnn_mfdfa_X_df.npy:\", rnn_mfdfa_X_df.npy.shape)\n",
     "\n",
     "# Convert the 3D CNN data to a DataFrame\n",
-    "cnn_X_df = pd.DataFrame(cnn_X_data)\n",
+    "cnn_X_df = pd.DataFrame(cnn_mfdfa_X_df)\n",
     "\n",
     "# Convert the 3D RNN data to a list of DataFrames with generic column labels\n",
-    "rnn_X_df = [pd.DataFrame(time_series) for time_series in rnn_X_data]\n",
+    "rnn_X_df = [pd.DataFrame(time_series) for time_series in rnn_mfdfa_X_df.npy]\n",
     "\n",
     "# Display the head of the first DataFrame in the list\n",
     "print(\"cnn_X_df:\")\n",
@@ -1720,7 +1749,7 @@
    "id": "2ed88397-ca95-433d-a081-ffb876bd61de",
    "metadata": {},
    "source": [
-    "# Save MFDFA RNN X data as DF"
+    "# Save MFDFA RNN X data and Hurst exp as DF"
    ]
   },
   {
@@ -1773,14 +1802,6 @@
     "rnn_data_path = '/home/vincent/AAA_projects/MVCS/Neuroscience/RNN_data/'\n",
     "np.save(f\"{rnn_data_path}rnn_X_data_combined.npy\", rnn_X_df_combined)"
    ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "ec327871-1f8d-43c3-bcef-dc941606890a",
-   "metadata": {},
-   "outputs": [],
-   "source": []
   }
  ],
  "metadata": {