diff --git a/napeca/bidi_correction_manual.ipynb b/napeca/bidi_correction_manual.ipynb
new file mode 100644
index 0000000..3314277
--- /dev/null
+++ b/napeca/bidi_correction_manual.ipynb
@@ -0,0 +1,143 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "import os\n",
+    "import h5py\n",
+    "import sima\n",
+    "from sima import sequence\n",
+    "from sima.imaging import ImagingDataset\n",
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "import bidi_offset_correction\n",
+    "import sima_motion_bidi_correction as bidi_funcs"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "fname = '22_1_19_gck_cgm_3_sacchari-001' # h5 file name (without the extension)\n",
+    "fdir = r'D:\\bruker_data\\Jennifer\\22_1_19_gck_cgm_3_sacchari-001' # root folder of h5 to cut\n",
+    "data_format = 'sima'\n",
+    "manual_offset_value = 10 # in pixels (I found this to be pixels you measure by eye +1)\n",
+    "\n",
+    "num_save_frames = 300 # use the following for saving and applying offset to all frames: data_n_meta['data'].shape[0] "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# load an object referencing the motion-corrected data\n",
+    "data_n_meta = bidi_funcs.load_sima_or_h5_object(fdir, fname, data_format)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# mean_subsample_img = np.squeeze(np.nanmean(data_n_meta['data'][::40,...], axis=0))\n",
+    "# my_bidi_corr_obj = bidi_offset_correction.bidi_offset_correction(mean_subsample_img)  # initialize data to object\n",
+    "\n",
+    "# my_bidi_corr_obj.compute_mean_image()  # compute mean image across time\n",
+    "# bidi_offset = my_bidi_corr_obj.determine_bidi_offset()  # calculated bidirectional offset via fft cross-correlation"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# applies an offset to odd lines. Every time you run this, it will update the sima offsets file!!!\n",
+    "bidi_funcs.apply_bidi_corr_to_sima_offsets(fdir, fname, -manual_offset_value-1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# show mean image\n",
+    "dataset = sima.ImagingDataset.load(os.path.join(fdir, os.path.splitext(fname)[0] + '_mc.sima'))\n",
+    "sequence_data = dataset.sequences[0][:num_save_frames,...]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# comment this out if you don't want to wait for the mean image to process and show in this notebook\n",
+    "plt.figure(figsize=(7,7))\n",
+    "plt.imshow(np.nanmean(np.squeeze(sequence_data), axis=0))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# saves an h5 based on the motion correction shifts calculated by sima\n",
+    "data_to_save = np.empty([num_save_frames, dataset.frame_shape[1], dataset.frame_shape[2]])\n",
+    "frame_iter1 = iter(sequence_data)\n",
+    "\n",
+    "# these next few lines just fill in the empty spaces\n",
+    "fill_gapscaller = bidi_funcs.fill_gaps(0, sequence_data, frame_iter1)\n",
+    "fill_gapscaller.send(None)\n",
+    "\n",
+    "for frame_num in range(num_save_frames):\n",
+    "    data_to_save[frame_num, ...] = fill_gapscaller.send(frame_num).astype('uint8')\n",
+    "\n",
+    "# save to an h5\n",
+    "sima_mc_bidi_outpath = os.path.join(fdir, fname + '_sima_mc.h5')\n",
+    "h5_write_bidi_corr = h5py.File(sima_mc_bidi_outpath, 'w')\n",
+    "h5_write_bidi_corr.create_dataset('imaging', data=data_to_save.astype('uint8'))\n",
+    "h5_write_bidi_corr.close()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 2",
+   "language": "python",
+   "name": "python2"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 2
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython2",
+   "version": "2.7.16"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}