autogaita 0.1.0__tar.gz → 0.2.0__tar.gz

autogaita-0.2.0/PKG-INFO

@@ -0,0 +1,10 @@
+Metadata-Version: 2.1
+Name: autogaita
+Version: 0.2.0
+Summary: Automatic Gait Analysis in Python
+Home-page: https://github.com/mahan-hosseini/AutoGaitA/
+Author: Mahan Hosseini
+License: GPLv3
+Requires-Python: >=3.10
+Provides-Extra: dev
+License-File: LICENSE

{autogaita-0.1.0 → autogaita-0.2.0}/README.md

@@ -16,7 +16,7 @@
 
 ## Getting Started
 
-***Note!** Our documentation provides step-by-step walkthroughs of how to install autogaita for **[Windows](https://docs.google.com/document/d/1Y4wrrsjs0ybLDKPzE2LAatqPDq9jtwjIuk4M0jRZ3wE/edit#heading=h.28j6wu2vamre)** and **[Mac](https://docs.google.com/document/d/1Y4wrrsjs0ybLDKPzE2LAatqPDq9jtwjIuk4M0jRZ3wE/edit)***
+***Note!** [Our documentation](https://docs.google.com/document/d/1Y4wrrsjs0ybLDKPzE2LAatqPDq9jtwjIuk4M0jRZ3wE/edit?usp=sharing) provides step-by-step walkthroughs of how to install autogaita for **[Windows](https://docs.google.com/document/d/1Y4wrrsjs0ybLDKPzE2LAatqPDq9jtwjIuk4M0jRZ3wE/edit#heading=h.28j6wu2vamre)** and **[Mac](https://docs.google.com/document/d/1Y4wrrsjs0ybLDKPzE2LAatqPDq9jtwjIuk4M0jRZ3wE/edit)***
 
 It is strongly recommended that a separate virtual environment for AutoGaitA is created (note that the approach below creates the virtual environment to your current directory):
 
@@ -31,6 +31,8 @@ It is strongly recommended that a separate virtual environment for AutoGaitA is
 
 - Access the main user interface via `python -m autogaita`.
 
+- To update to the latest release (see the *Releases* panel on the right for the current version) activate your virtual environment & enter `pip install autogaita -U`. 
+
 ## Tutorials and Examples
 
 ### Video Walkthrough Tutorials
@@ -45,7 +47,7 @@ We provide an example dataset in the **example data** folder of this repository,
 ### Annotation Table Examples and Templates
 Annotation Table example and template files for *AutoGaitA_DLC* and *AutoGaitA_Simi* can be found in the [**annotation tables**](https://github.com/mahan-hosseini/AutoGaitA/tree/main/annotation%20tables) folder of this repository.
 
-Users are advised to read the ***important note*** of that folder, use the template to enter their data's timestamp information and to then compare the resulting table with our example to check formatting.
+Users are advised to read the ***important note*** of that folder, use the template to enter their data's timestamp information and to then compare the resulting table with our example to check formatting. Users working with ImageJ/FIJI are encouraged to check out the [AnnotationTable-Plugin](https://github.com/luca-flemming/AnnotationTable-Plugin) developed by our contributor Luca Flemming.
 
 ## Documentation
 
@@ -62,6 +64,9 @@ By default, *AutoGaitA DLC* and *AutoGaitA Simi* implement standard values for m
 - [Documentation - AutoGaitA DLC](https://docs.google.com/document/d/1Y4wrrsjs0ybLDKPzE2LAatqPDq9jtwjIuk4M0jRZ3wE/edit#heading=h.20bg7b7ymt0b)
 - [Documentation - AutoGaitA Simi](https://docs.google.com/document/d/1Y4wrrsjs0ybLDKPzE2LAatqPDq9jtwjIuk4M0jRZ3wE/edit#heading=h.uz61bpmua7qz)
 
+## Analysing other behaviours - AutoCyclA 🚴
+Even though AutoGaitA's main focus is to automate and standardise gait analyses, our toolbox can be used to automate the analyses of any rhythmic behaviour of interest. For a proof-of-principle demonstration and an introduction of the general workflow of such analyses, see **[AutoCyclA - Automated Cycling Analysis with AutoGaitA.](https://github.com/mahan-hosseini/AutoGaitA/tree/main/autocycla)**
+
 ## Reference
 If you use this code or data please [cite our preprint](https://www.biorxiv.org/content/10.1101/2024.04.14.589409v1).
 
@@ -71,4 +76,14 @@ AutoGaitA is licensed under [GPL v3.0](https://github.com/mahan-hosseini/AutoGai
 The AutoGaitA software is provided without warranty of any kind, express or implied, including, but not limited to, the implied warranty of fitness for a particular purpose.
 
 ## Authors
-Mahan Hosseini
+[Mahan Hosseini](https://github.com/mahan-hosseini)
+
+## Contributors
+[Luca Flemming](https://github.com/luca-flemming) - Undergraduate Student
+
+[Nicholas del Grosso](https://github.com/nickdelgrosso) - RSE Advisor
+
+## Contributing
+If you would like to contribute to the AutoGaitA toolbox, feel free to open a pull request or contact us at autogaita@fz-juelich.de! 
+
+We are looking forward to your input and ideas 😊

{autogaita-0.1.0 → autogaita-0.2.0}/autogaita/autogaita_dlc.py

@@ -9,6 +9,7 @@ import numpy as np
 import math
 import matplotlib
 import matplotlib.pyplot as plt
+import seaborn as sns
 
 # %% constants
 plt.rcParams["figure.dpi"] = 300  # increase resolution of figures
@@ -38,6 +39,7 @@ AVERAGE_XLS_FILENAME = " - Average Stepcycle"
 STD_XLS_FILENAME = " - Standard Devs. Stepcycle"
 SC_LAT_LEGEND_FONTSIZE = 7
 
+
 # %% main program
 
 
@@ -92,6 +94,7 @@ def some_prep(info, folderinfo, cfg):
     normalise_height_at_SC_level = cfg["normalise_height_at_SC_level"]
     invert_y_axis = cfg["invert_y_axis"]
     flip_gait_direction = cfg["flip_gait_direction"]
+    export_average_x = cfg["export_average_x"]
 
     # .............................  move data  ........................................
     # => see if we can delete a previous runs results folder if existant. if not, it's a
@@ -127,8 +130,8 @@ def some_prep(info, folderinfo, cfg):
             "\n******************\n! CRITICAL ERROR !\n******************\n"
             + "Unable to identify ANY RELEVANT FILES for "
             + name
-            + "!\nThis is likely due to issues with pre/post-strings.. check "
-            + "capitalisation!"
+            + "!\nThis is likely due to issues with unique file name identifiers.. "
+            + "check capitalisation!"
         )
         write_issues_to_textfile(no_files_error, info)
         print(no_files_error)
@@ -138,6 +141,15 @@ def some_prep(info, folderinfo, cfg):
     datadf = pd.DataFrame(data=None)  # prep stuff for error handling
     datadf_duplicate_error = ""
     if subtract_beam:
+        if data_string == beam_string:
+            beam_and_data_string_error_message = (
+                "\n******************\n! CRITICAL ERROR !\n******************\n"
+                + "Your data & baseline (beam) identifiers ([G] in our "
+                + "file  naming convention) are identical. "
+                + "\nNote that they must be different! \nTry again"
+            )
+            write_issues_to_textfile(beam_and_data_string_error_message, info)
+            return
         beamdf = pd.DataFrame(data=None)
         beamdf_duplicate_error = ""
     for filename in os.listdir(results_dir):  # import
@@ -216,7 +228,7 @@ def some_prep(info, folderinfo, cfg):
     # IMPORTANT
     # ---------
     # MAIN TESTS OF USER-INPUT VALIDITY OCCUR HERE!
-    cfg = test_and_expand_cfg(data, cfg, info)
+    cfg = check_and_expand_cfg(data, cfg, info)
     if cfg is None:  # hind joints were empty
         return
     hind_joints = cfg["hind_joints"]
@@ -225,7 +237,6 @@ def some_prep(info, folderinfo, cfg):
     beam_hind_jointadd = cfg["beam_hind_jointadd"]
     beam_fore_jointadd = cfg["beam_fore_jointadd"]
     direction_joint = cfg["direction_joint"]
-    export_average_x = cfg["export_average_x"]
     # store config json file @ group path
     # !!! NU - do this @ mouse path!
     group_path = results_dir.split(name)[0]
@@ -243,6 +254,29 @@ def some_prep(info, folderinfo, cfg):
     # note - using "w" will overwrite/truncate file, thus no need to remove it if exists
     with open(config_json_path, "w") as config_json_file:
         json.dump(config_vars_to_json, config_json_file, indent=4)
+    # a little test to see if columns make sense, i.e., same number of x/y/likelihood
+    x_col_count = len([c for c in data.columns if c.endswith(" x")])
+    y_col_count = len([c for c in data.columns if c.endswith(" y")])
+    likelihood_col_count = len([c for c in data.columns if c.endswith(" likelihood")])
+    if x_col_count == y_col_count == likelihood_col_count:
+        pass
+    else:
+        cols_are_weird_message = (
+            "\n***********\n! WARNING !\n***********\n"
+            + "We detected an unequal number of columns ending with x, y or "
+            + "likelihood!\nCounts were:\n"
+            + "x: "
+            + str(x_col_count)
+            + ", y: "
+            + str(y_col_count)
+            + ", likelihood: "
+            + str(likelihood_col_count)
+            + "!\n\n"
+            + "We continue with the analysis but we strongly suggest you have another "
+            + "look at your dataset, this should not happen.\n"
+        )
+        print(cols_are_weird_message)
+        write_issues_to_textfile(cols_are_weird_message, info)
     # if wanted: fix that deeplabcut inverses y
     if invert_y_axis:
         for col in data.columns:
@@ -346,7 +380,7 @@ def move_data_to_folders(info, folderinfo):
                 write_issues_to_textfile(this_message, info)
 
 
-def test_and_expand_cfg(data, cfg, info):
+def check_and_expand_cfg(data, cfg, info):
     """Test some important cfg variables and add new ones based on them
 
     Procedure
@@ -402,7 +436,7 @@ def test_and_expand_cfg(data, cfg, info):
         return
     cfg["direction_joint"] = hind_joints[0]
 
-    # if subtracting beam, check that its colnames were valid.
+    # if subtracting beam, check identifier-strings & that beam colnames were valid.
     if cfg["subtract_beam"]:
         beam_col_error_message = (
             "\n******************\n! CRITICAL ERROR !\n******************\n"
@@ -585,12 +619,10 @@ def flip_mouse_body(data, info):
     write_issues_to_textfile(message, info)
 
     # 1) Flip all rows in x columns only and subtract max from all vals
-    flipped_data = pd.DataFrame(data=None, columns=data.columns)
-    for col in data.columns:
-        if col.endswith("x"):
-            flipped_data.loc[:, col] = max(data.loc[:, col]) - data.loc[:, col]
-        else:
-            flipped_data.loc[:, col] = data.loc[:, col]
+    flipped_data = data.copy()
+    x_cols = [col for col in flipped_data.columns if col.endswith(" x")]
+    for col in x_cols:
+        flipped_data[col] = max(flipped_data[col]) - flipped_data[col]
     return flipped_data
 
 
@@ -1330,6 +1362,24 @@ def add_step_separators(dataframe, nanvector, numvector):
 
 
 # ..............................  master function  .............................
+
+# A note on updated colour cyclers after pull request that was merged 20.06.2024
+# => Using color palettes instead of colour maps as I had previously means that
+#    we cycle through neighbouring colours
+# => I initially implemented an "equally distant" approach.
+# => So for viridis and n=2 (e.g. if 2 groups) it would be purple and yellow
+#    (as far away as possible)
+# => Now it is dark blue and green
+# => Updated approach is aesthetically more pleasing IMO.
+# => However it does have the risk of not being able to tell the colours in
+#    some cases - e.g. if some accelerations are very overlapping.
+# => But - because users can in theses cases just choose a colour palette that
+#    in itself has categorical colours (Set1, Dark2, etc.) I still keep the new
+#    behaviour
+# => Nonetheless, in case you want to use the "old behaviour" at some point it
+#    would be coded as commented out in plot_joint_y_by_x
+
+
 def plot_results(info, results, folderinfo, cfg):
     """Plot results - y coords by x coords & average angles over SC %"""
     # unpack
@@ -1439,6 +1489,8 @@ def plot_joint_y_by_x(all_steps_data, sc_idxs, info, cfg):
     convert_to_mm = cfg["convert_to_mm"]
     plot_joints = cfg["plot_joints"]
     sampling_rate = cfg["sampling_rate"]
+    legend_outside = cfg["legend_outside"]
+    color_palette = cfg["color_palette"]
 
     # some prep
     sc_num = len(sc_idxs)
@@ -1448,49 +1500,37 @@ def plot_joint_y_by_x(all_steps_data, sc_idxs, info, cfg):
     # plot
     for j, joint in enumerate(plot_joints):  # joint loop (figures)
         f[j], ax[j] = plt.subplots(1, 1)
-        ax[j].set_prop_cycle(
-            plt.cycler("color", plt.cm.viridis(np.linspace(0, 1, sc_num)))
+        # What "Old" colormap approach would look like with seaborn
+        # this_map = sns.color_palette(cfg["color_palette"], as_cmap=True)
+        # ax[j].set_prop_cycle(plt.cycler("color", this_map(np.linspace(0, 1, sc_num))))
+        ax[j].set_prop_cycle(  # New color palette approach
+            plt.cycler("color", sns.color_palette(color_palette, sc_num))
         )
-        if joint == "Hind paw tao ":
-            ax[j].set_title(name + " - Foot")
-        else:
-            ax[j].set_title(name + " - " + joint)
+        ax[j].set_title(name + " - " + joint)
         x_col_idx = all_steps_data.columns.get_loc(joint + "x")
         y_col_idx = all_steps_data.columns.get_loc(joint + "y")
         time_col_idx = all_steps_data.columns.get_loc(TIME_COL)
         for s in range(sc_num):
             this_x = all_steps_data.iloc[sc_idxs[s], x_col_idx]
             this_y = all_steps_data.iloc[sc_idxs[s], y_col_idx]
-            if sampling_rate <= 100:
-                float_precision = 2  # how many decimals we round to
-            else:
-                float_precision = 4
-            this_label = (
-                str(
-                    round(
-                        all_steps_data.iloc[sc_idxs[s][0], time_col_idx],
-                        float_precision,
-                    )
-                )
-                + "-"
-                + str(
-                    round(
-                        all_steps_data.iloc[sc_idxs[s][-1], time_col_idx],
-                        float_precision,
-                    )
-                )
-                + "s"
+            this_label = generate_sc_latency_label(
+                all_steps_data, sc_idxs[s], sampling_rate, time_col_idx
             )
             ax[j].plot(this_x, this_y, label=this_label)
         ax[j].set_xlabel("x (pixel)")  # will be overwritten if we convert
         ax[j].set_ylabel("y (pixel)")
-        ax[j].legend(fontsize=SC_LAT_LEGEND_FONTSIZE)
+        # legend adjustments
+        if legend_outside is True:
+            ax[j].legend(
+                fontsize=SC_LAT_LEGEND_FONTSIZE,
+                loc="center left",
+                bbox_to_anchor=(1, 0.5),
+            )
+        elif legend_outside is False:
+            ax[j].legend(fontsize=SC_LAT_LEGEND_FONTSIZE)
         if convert_to_mm:
             tickconvert_mm_to_cm(ax[j], "both")
-        if joint == "Hind paw tao ":
-            figure_file_string = " - Foot y by x coordinates"
-        else:
-            figure_file_string = " - " + joint + "y by x coordinates"
+        figure_file_string = " - " + joint + "y by x coordinates"
         save_figures(f[j], results_dir, name, figure_file_string)
         if dont_show_plots:
             plt.close(f[j])
@@ -1504,6 +1544,9 @@ def plot_angles_by_time(all_steps_data, sc_idxs, info, cfg):
     results_dir = info["results_dir"]
     dont_show_plots = cfg["dont_show_plots"]
     angles = cfg["angles"]
+    sampling_rate = cfg["sampling_rate"]
+    legend_outside = cfg["legend_outside"]
+    color_palette = cfg["color_palette"]
 
     # some prep
     sc_num = len(sc_idxs)
@@ -1514,17 +1557,30 @@ def plot_angles_by_time(all_steps_data, sc_idxs, info, cfg):
     for a, angle in enumerate(angles["name"]):  # angle loop (figures)
         f[a], ax[a] = plt.subplots(1, 1)
         ax[a].set_prop_cycle(
-            plt.cycler("color", plt.cm.viridis(np.linspace(0, 1, sc_num)))
+            plt.cycler("color", sns.color_palette(color_palette, sc_num))
         )
         ax[a].set_title(name + " - " + angle)
         ax[a].set_ylabel("Angle")
         ax[a].set_xlabel("Time (s)")
         x_col_idx = all_steps_data.columns.get_loc(TIME_COL)
         y_col_idx = all_steps_data.columns.get_loc(angle + "Angle")
+        time_col_idx = all_steps_data.columns.get_loc(TIME_COL)
         for s in range(sc_num):
             this_x = all_steps_data.iloc[sc_idxs[s], x_col_idx]
             this_y = all_steps_data.iloc[sc_idxs[s], y_col_idx]
-            ax[a].plot(this_x, this_y)
+            this_label = generate_sc_latency_label(
+                all_steps_data, sc_idxs[s], sampling_rate, time_col_idx
+            )
+            ax[a].plot(this_x, this_y, label=this_label)
+        # legend adjustments
+        if legend_outside is True:
+            ax[a].legend(
+                fontsize=SC_LAT_LEGEND_FONTSIZE,
+                loc="center left",
+                bbox_to_anchor=(1, 0.5),
+            )
+        elif legend_outside is False:
+            ax[a].legend(fontsize=SC_LAT_LEGEND_FONTSIZE)
         figure_file_string = " - " + angle + "Angle by Time"
         save_figures(f[a], results_dir, name, figure_file_string)
         if dont_show_plots:
@@ -1541,11 +1597,13 @@ def plot_hindlimb_stickdiagram(all_steps_data, sc_idxs, info, cfg):
     convert_to_mm = cfg["convert_to_mm"]
     plot_joints = cfg["plot_joints"]
     sampling_rate = cfg["sampling_rate"]
+    legend_outside = cfg["legend_outside"]
+    color_palette = cfg["color_palette"]
 
     # some prep
     sc_num = len(sc_idxs)
     f, ax = plt.subplots(1, 1)
-    color_cycle = plt.cycler("color", plt.cm.viridis(np.linspace(0, 1, sc_num)))
+    color_cycle = plt.cycler("color", sns.color_palette(color_palette, sc_num))
     ax.set_prop_cycle(color_cycle)
     time_col_idx = all_steps_data.columns.get_loc(TIME_COL)
 
@@ -1553,21 +1611,8 @@ def plot_hindlimb_stickdiagram(all_steps_data, sc_idxs, info, cfg):
     # => for timepoints from SC1 to SCend - plot(joint1x, joint1y)
     for s, this_color_dict in zip(range(sc_num), color_cycle):  # SC loop (colors)
         this_color = this_color_dict["color"][:3]
-        if sampling_rate <= 100:
-            float_precision = 2  # how many decimals we round to
-        else:
-            float_precision = 4
-        this_label = (
-            str(
-                round(all_steps_data.iloc[sc_idxs[s][0], time_col_idx], float_precision)
-            )
-            + "-"
-            + str(
-                round(
-                    all_steps_data.iloc[sc_idxs[s][-1], time_col_idx], float_precision
-                )
-            )
-            + "s"
+        this_label = generate_sc_latency_label(
+            all_steps_data, sc_idxs[s], sampling_rate, time_col_idx
         )
         for i in sc_idxs[s]:  # loop over timepoints of current SC
             this_xs = list()  # for each timepoint, define joints' xy coord new
@@ -1586,7 +1631,13 @@ def plot_hindlimb_stickdiagram(all_steps_data, sc_idxs, info, cfg):
     ax.set_ylabel("y (pixel)")
     if convert_to_mm:
         tickconvert_mm_to_cm(ax, "both")
-    ax.legend(fontsize=SC_LAT_LEGEND_FONTSIZE)
+    # legend adjustments
+    if legend_outside is True:
+        ax.legend(
+            fontsize=SC_LAT_LEGEND_FONTSIZE, loc="center left", bbox_to_anchor=(1, 0.5)
+        )
+    elif legend_outside is False:
+        ax.legend(fontsize=SC_LAT_LEGEND_FONTSIZE)
     figure_file_string = " - Hindlimb Stick Diagram"
     save_figures(f, results_dir, name, figure_file_string)
     if dont_show_plots:
@@ -1603,32 +1654,21 @@ def plot_forelimb_stickdiagram(all_steps_data, sc_idxs, info, cfg):
     convert_to_mm = cfg["convert_to_mm"]
     fore_joints = cfg["fore_joints"]
     sampling_rate = cfg["sampling_rate"]
+    legend_outside = cfg["legend_outside"]
+    color_palette = cfg["color_palette"]
 
     # some prep
     sc_num = len(sc_idxs)
     f, ax = plt.subplots(1, 1)
-    color_cycle = plt.cycler("color", plt.cm.viridis(np.linspace(0, 1, sc_num)))
+    color_cycle = plt.cycler("color", sns.color_palette(color_palette, sc_num))
     ax.set_prop_cycle(color_cycle)
     time_col_idx = all_steps_data.columns.get_loc(TIME_COL)
 
     # plot
     for s, this_color in zip(range(sc_num), color_cycle):  # SC loop (colors)
         this_color = this_color["color"][:3]
-        if sampling_rate <= 100:
-            float_precision = 2  # how many decimals we round to
-        else:
-            float_precision = 4
-        this_label = (
-            str(
-                round(all_steps_data.iloc[sc_idxs[s][0], time_col_idx], float_precision)
-            )
-            + "-"
-            + str(
-                round(
-                    all_steps_data.iloc[sc_idxs[s][-1], time_col_idx], float_precision
-                )
-            )
-            + "s"
+        this_label = generate_sc_latency_label(
+            all_steps_data, sc_idxs[s], sampling_rate, time_col_idx
         )
         for i in sc_idxs[s]:
             this_xs = list()
@@ -1647,7 +1687,13 @@ def plot_forelimb_stickdiagram(all_steps_data, sc_idxs, info, cfg):
     ax.set_ylabel("y (pixel)")
     if convert_to_mm:
         tickconvert_mm_to_cm(ax, "both")
-    ax.legend(fontsize=SC_LAT_LEGEND_FONTSIZE)
+    # legend adjustments
+    if legend_outside is True:
+        ax.legend(
+            fontsize=SC_LAT_LEGEND_FONTSIZE, loc="center left", bbox_to_anchor=(1, 0.5)
+        )
+    elif legend_outside is False:
+        ax.legend(fontsize=SC_LAT_LEGEND_FONTSIZE)
     figure_file_string = " - Forelimb Stick Diagram"
     save_figures(f, results_dir, name, figure_file_string)
     if dont_show_plots:
@@ -1666,11 +1712,13 @@ def plot_joint_y_by_average_SC(average_data, std_data, info, cfg):
     plot_SE = cfg["plot_SE"]
     sc_num = cfg["sc_num"]
     hind_joints = cfg["hind_joints"]
+    legend_outside = cfg["legend_outside"]
+    color_palette = cfg["color_palette"]
 
     # plot
     f, ax = plt.subplots(1, 1)
     ax.set_prop_cycle(
-        plt.cycler("color", plt.cm.viridis(np.linspace(0, 1, len(hind_joints))))
+        plt.cycler("color", sns.color_palette(color_palette, len(hind_joints)))
     )
     x = np.linspace(0, 100, bin_num)
     for joint in hind_joints:  # joint loop (lines)
@@ -1682,7 +1730,11 @@ def plot_joint_y_by_average_SC(average_data, std_data, info, cfg):
             this_std = std_data.iloc[:, y_col_idx]
         ax.plot(x, this_y, label=joint)
         ax.fill_between(x, this_y - this_std, this_y + this_std, alpha=0.2)
-    ax.legend()
+    # legend adjustments
+    if legend_outside is True:
+        ax.legend(loc="center left", bbox_to_anchor=(1, 0.5))
+    elif legend_outside is False:
+        ax.legend()
     ax.set_title(name + " - Joint Y over average step cycle")
     ax.set_xlabel("Percentage")
     ax.set_ylabel("y (pixel)")
@@ -1705,11 +1757,13 @@ def plot_angles_by_average_SC(average_data, std_data, info, cfg):
     plot_SE = cfg["plot_SE"]
     sc_num = cfg["sc_num"]
     angles = cfg["angles"]
+    legend_outside = cfg["legend_outside"]
+    color_palette = cfg["color_palette"]
 
     # plot
     f, ax = plt.subplots(1, 1)
     ax.set_prop_cycle(
-        plt.cycler("color", plt.cm.viridis(np.linspace(0, 1, len(angles["name"]))))
+        plt.cycler("color", sns.color_palette(color_palette, len(angles["name"])))
     )
     x = np.linspace(0, 100, bin_num)
     ax.set_title(name + " - Joint angles over average step cycle")
@@ -1724,7 +1778,11 @@ def plot_angles_by_average_SC(average_data, std_data, info, cfg):
             this_std = std_data.iloc[:, y_col_idx]
         ax.plot(x, this_y, label=angle)
         ax.fill_between(x, this_y - this_std, this_y + this_std, alpha=0.2)
-    ax.legend()
+    # legend adjustments
+    if legend_outside is True:
+        ax.legend(loc="center left", bbox_to_anchor=(1, 0.5))
+    elif legend_outside is False:
+        ax.legend()
     figure_file_string = " - Joint angles over average step cycle"
     save_figures(f, results_dir, name, figure_file_string)
     if dont_show_plots:
@@ -1744,11 +1802,13 @@ def plot_x_velocities_by_average_SC(average_data, std_data, info, cfg):
     plot_SE = cfg["plot_SE"]
     sc_num = cfg["sc_num"]
     hind_joints = cfg["hind_joints"]
+    legend_outside = cfg["legend_outside"]
+    color_palette = cfg["color_palette"]
 
     # plot
     f, ax = plt.subplots(1, 1)
     ax.set_prop_cycle(
-        plt.cycler("color", plt.cm.viridis(np.linspace(0, 1, len(hind_joints))))
+        plt.cycler("color", sns.color_palette(color_palette, len(hind_joints)))
     )
     x = np.linspace(0, 100, bin_num)
     ax.set_title(name + " - Joint velocities over average step cycle")
@@ -1761,7 +1821,11 @@ def plot_x_velocities_by_average_SC(average_data, std_data, info, cfg):
             this_std = std_data.iloc[:, y_col_idx]
         ax.plot(x, this_y, label=joint)
         ax.fill_between(x, this_y - this_std, this_y + this_std, alpha=0.2)
-    ax.legend()
+    # legend adjustments
+    if legend_outside is True:
+        ax.legend(loc="center left", bbox_to_anchor=(1, 0.5))
+    elif legend_outside is False:
+        ax.legend()
     ax.set_xlabel("Percentage")
     ax.set_ylabel(
         "Velocity (x in pixel / " + str(int((1 / sampling_rate) * 1000)) + "ms)"
@@ -1789,11 +1853,13 @@ def plot_angular_velocities_by_average_SC(average_data, std_data, info, cfg):
     plot_SE = cfg["plot_SE"]
     sc_num = cfg["sc_num"]
     angles = cfg["angles"]
+    legend_outside = cfg["legend_outside"]
+    color_palette = cfg["color_palette"]
 
     # plot
     f, ax = plt.subplots(1, 1)
     ax.set_prop_cycle(
-        plt.cycler("color", plt.cm.viridis(np.linspace(0, 1, len(angles["name"]))))
+        plt.cycler("color", sns.color_palette(color_palette, len(angles["name"])))
     )
     x = np.linspace(0, 100, bin_num)
     ax.set_title(name + " - Angular velocities over average step cycle")
@@ -1808,7 +1874,11 @@ def plot_angular_velocities_by_average_SC(average_data, std_data, info, cfg):
             this_std = std_data.iloc[:, y_col_idx]
         ax.plot(x, this_y, label=angle)
         ax.fill_between(x, this_y - this_std, this_y + this_std, alpha=0.2)
-    ax.legend()
+    # legend adjustments
+    if legend_outside is True:
+        ax.legend(loc="center left", bbox_to_anchor=(1, 0.5))
+    elif legend_outside is False:
+        ax.legend()
     figure_file_string = " - Angular velocities over average step cycle"
     save_figures(f, results_dir, name, figure_file_string)
     if dont_show_plots:
@@ -1828,11 +1898,13 @@ def plot_x_acceleration_by_average_SC(average_data, std_data, info, cfg):
     plot_SE = cfg["plot_SE"]
     sc_num = cfg["sc_num"]
     hind_joints = cfg["hind_joints"]
+    legend_outside = cfg["legend_outside"]
+    color_palette = cfg["color_palette"]
 
     # plot
     f, ax = plt.subplots(1, 1)
     ax.set_prop_cycle(
-        plt.cycler("color", plt.cm.viridis(np.linspace(0, 1, len(hind_joints))))
+        plt.cycler("color", sns.color_palette(color_palette, len(hind_joints)))
     )
     x = np.linspace(0, 100, bin_num)
     ax.set_title(name + " - Joint accelerations over average step cycle")
@@ -1845,7 +1917,11 @@ def plot_x_acceleration_by_average_SC(average_data, std_data, info, cfg):
             this_std = std_data.iloc[:, y_col_idx]
         ax.plot(x, this_y, label=joint)
         ax.fill_between(x, this_y - this_std, this_y + this_std, alpha=0.2)
-    ax.legend()
+    # legend adjustments
+    if legend_outside is True:
+        ax.legend(loc="center left", bbox_to_anchor=(1, 0.5))
+    elif legend_outside is False:
+        ax.legend()
     ax.set_xlabel("Percentage")
     ax.set_ylabel(
         "Acceleration (x in pixel / " + str(int((1 / sampling_rate) * 1000)) + "ms)"
@@ -1875,11 +1951,13 @@ def plot_angular_acceleration_by_average_SC(average_data, std_data, info, cfg):
     plot_SE = cfg["plot_SE"]
     sc_num = cfg["sc_num"]
     angles = cfg["angles"]
+    legend_outside = cfg["legend_outside"]
+    color_palette = cfg["color_palette"]
 
     # plot
     f, ax = plt.subplots(1, 1)
     ax.set_prop_cycle(
-        plt.cycler("color", plt.cm.viridis(np.linspace(0, 1, len(angles["name"]))))
+        plt.cycler("color", sns.color_palette(color_palette, len(angles["name"])))
     )
     x = np.linspace(0, 100, bin_num)
     ax.set_title(name + " - Angular accelerations over average step cycle")
@@ -1896,7 +1974,11 @@ def plot_angular_acceleration_by_average_SC(average_data, std_data, info, cfg):
             this_std = std_data.iloc[:, y_col_idx]
         ax.plot(x, this_y, label=angle)
         ax.fill_between(x, this_y - this_std, this_y + this_std, alpha=0.2)
-    ax.legend()
+    # legend adjustments
+    if legend_outside is True:
+        ax.legend(loc="center left", bbox_to_anchor=(1, 0.5))
+    elif legend_outside is False:
+        ax.legend()
     figure_file_string = " - Angular acceleration over average step cycle"
     save_figures(f, results_dir, name, figure_file_string)
     if dont_show_plots:
@@ -1937,6 +2019,30 @@ def tickconvert_mm_to_cm(axis, whichlabel):
         axis.set_ylabel("y (cm)")
 
 
+def generate_sc_latency_label(all_steps_data, this_sc_idx, sampling_rate, time_col_idx):
+    if sampling_rate <= 100:
+        float_precision = 2  # how many decimals we round to
+    else:
+        float_precision = 4
+    this_label = (
+        str(
+            round(
+                all_steps_data.iloc[this_sc_idx[0], time_col_idx],
+                float_precision,
+            )
+        )
+        + "-"
+        + str(
+            round(
+                all_steps_data.iloc[this_sc_idx[-1], time_col_idx],
+                float_precision,
+            )
+        )
+        + "s"
+    )
+    return this_label
+
+
 # %% local functions 5 - print finish