Skip to content

Commit

Permalink
Fixed combine baseline data
Browse files Browse the repository at this point in the history 
 * All baseline data (pre and post) can be combined with the current code.
* ToDo :
  * Experimental data is not fixed yet
  • Loading branch information
@ihgumilar
ihgumilar committed Sep 1, 2022
1 parent 964d1fa commit 201ad99
Showing 1 changed file with 105 additions and 191 deletions.
296 changes: 105 additions & 191 deletions EyeTracker/pre-processing/combine_2_csv_files_eye_tracker_data.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -26,284 +26,198 @@
# ### Baseline data

# %% [markdown]
# #### Define a directory where we store raw eeg files (*.fif), which are not combined yet (Baseline data)
# #### Define some directories :
# * Where we store raw eeg files (*.fif), which are not combined yet (Baseline data)
# * Where we would like to store combined eye tracker data (Baseline data)
# * Where we would like to store combined eye tracker data (Experiment data)
#

# %%
# Go to a directory that stores raw fif file (not combined files)
# Raw baseline data (not combined yet)
raw_dir = "/hpc/igum002/codes/Hyperscanning2-redesign/data/EyeTracker/raw_baseline_eye_data/"
raw_combined_baseline_data_dir = "/hpc/igum002/codes/Hyperscanning2-redesign/data/EyeTracker/raw_baseline_eye_data/raw_combined_baseline_eye_data/"
raw_combined_experimental_data_dir = "/hpc/igum002/codes/Hyperscanning2-redesign/data/EyeTracker/raw_experimental_eye_data/raw_combined_experimental_eye_data/"

os.chdir(raw_dir)

# %%
#TODO : Delete this temporary
data1 = {'Name':['Tom', 'Brad', 'Kyle', 'Jerry'],
'Age':[20, 21, 19, 18],
'Height' : [6.1, 5.9, 6.0, 6.1]
}

data2 = {'Name':['Tom', 'Brad', 'Kyle', 'Jerry'],
'Age':[30, 40, 50, 60],
'Height' : [7, 8, 9, 10]
}

df1 = pd.DataFrame(data1)
df2 = pd.DataFrame(data2)
print(df1.head())
print()
print(df2.head())
# Folder to store combined eye tracker data (baseline data)
raw_combined_baseline_data_directory = "/hpc/igum002/codes/Hyperscanning2-redesign/data/EyeTracker/raw_baseline_eye_data/raw_combined_baseline_eye_data/"
# Folder to store combined eye tracker data (experimental data)
raw_combined_experimental_data_directory = "/hpc/igum002/codes/Hyperscanning2-redesign/data/EyeTracker/raw_experimental_eye_data/raw_combined_experimental_eye_data/"

# %%
df = pd.concat([df1, df2])
df

# %% [markdown]
# #### Combine pre averted baseline

# %%
for i in tqdm(range(16), desc="Combining pre averted..."): # type: ignore
# Pre-averted
for i in tqdm(range(15), desc="Combining pre averted..."): # type: ignore
# Pre-averted (for processing subject 1 - 9)
if i < 9:

# FIXME : Load using pandas
# averted_pre_right_odd_subject = mne.io.read_raw_fif(
# "EyeTracker-S0" + str(i + 1) + "-averted_pre_right_point_raw.csv", verbose=False
# )
# TODO : In progress. Keep continuing and testing
averted_pre_right_odd_subject_file_name = f"EyeTracker-S0 + str(i + 1) + -averted_pre_right_point_raw.csv"
# Load averted pre right
averted_pre_right_odd_subject_file_name = "EyeTracker-S0" + str(i + 1) + "-averted_pre_right_point_raw.csv"
averted_pre_right_odd_subject = pd.read_csv(averted_pre_right_odd_subject_file_name)

# FIXME : Load using pandas
# averted_pre_left_odd_subject = mne.io.read_raw_fif(
# "EyeTracker-S0" + str(i + 1) + "-averted_pre_left_point_raw.csv", verbose=False
# )

# TODO : In progress. Keep continuing and testing
averted_pre_left_odd_subject_file_name = f"EyeTracker-S0" + str(i + 1) + "-averted_pre_left_point_raw.csv"
# Load Load averted pre left
averted_pre_left_odd_subject_file_name = "EyeTracker-S0" + str(i + 1) + "-averted_pre_left_point_raw.csv"
averted_pre_left_odd_subject = pd.read_csv(averted_pre_left_odd_subject_file_name)

# Check if i (index in this looping) == EVEN number that takes ODD actual subject no, then labeling the file name of hand is RIGHT-LEFT
# For example, i = 0 actually takes S01 and keeps going...
if (i % 2) == 0:

averted_pre_files_to_combine = [
averted_pre_right_odd_subject,
averted_pre_left_odd_subject,
]
combined_pre_averted_files = mne.concatenate_raws(
averted_pre_files_to_combine
)

# Combine RIGHT => LEFT hand data
combined_pre_averted_files = pd.concat([averted_pre_right_odd_subject, averted_pre_left_odd_subject])

# Create file name for combine files of pre-averted baseline
combined_pre_averted_files_label = (
raw_combined_baseline_data_dir
+ "S0"
+ str(i + 1)
+ "-averted_pre_right_left_point_combined_raw.csv"
raw_combined_baseline_data_directory + "S0" + str(i + 1) + "-averted_pre_right_left_point_combined_raw.csv"
)

# Save combine pre-averted baseline file to csv
combined_pre_averted_files.to_csv(combined_pre_averted_files_label)


# Check if i (index in this looping) == ODD number that takes EVEN actual subject no, then labeling the file name of hand is LEFT_RIGHT
# For example, i = 1 actually takes S02 and keeps going...

else:

averted_pre_files_to_combine = [
averted_pre_left_odd_subject,
averted_pre_right_odd_subject,
]

combined_pre_averted_files = mne.concatenate_raws(
averted_pre_files_to_combine
)

# Combine LEFT => RIGHT hand data
combined_pre_averted_files = pd.concat([averted_pre_left_odd_subject, averted_pre_right_odd_subject])

# Create file name for combine files of pre-averted baseline
combined_pre_averted_files_label = (
raw_combined_baseline_data_dir
+ "S0"
+ str(i + 1)
+ "-averted_pre_left_right_point_combined_raw.csv"
raw_combined_baseline_data_directory + "S0" + str(i + 1) + "-averted_pre_left_right_point_combined_raw.csv"
)

combined_pre_averted_files.save( # type: ignore
combined_pre_averted_files_label, overwrite=True
)
# Save combine pre-averted baseline file to csv
combined_pre_averted_files.to_csv(combined_pre_averted_files_label)

# Pre-averted (for processing subject 10 onwards)
else:
# Load averted pre right
averted_pre_right_odd_subject_file_name = "EyeTracker-S" + str(i + 1) + "-averted_pre_right_point_raw.csv"
averted_pre_right_odd_subject = pd.read_csv(averted_pre_right_odd_subject_file_name)

averted_pre_right_odd_subject = mne.io.read_raw_fif(
"EyeTracker-S" + str(i + 1) + "-averted_pre_right_point_raw.csv", verbose=False
)
averted_pre_left_odd_subject = mne.io.read_raw_fif(
"EyeTracker-S" + str(i + 1) + "-averted_pre_left_point_raw.csv", verbose=False
)
averted_pre_files_to_combine = [
averted_pre_right_odd_subject,
averted_pre_left_odd_subject,
]
combined_pre_averted_files = mne.concatenate_raws(averted_pre_files_to_combine)
# Load Load averted pre left
averted_pre_left_odd_subject_file_name = "EyeTracker-S" + str(i + 1) + "-averted_pre_left_point_raw.csv"
averted_pre_left_odd_subject = pd.read_csv(averted_pre_left_odd_subject_file_name)

# Check if i (index in this looping) == EVEN number that takes ODD actual subject no, then labeling the file name of hand is RIGHT-LEFT
# For example, i = 0 actually takes S01 and keeps going...
if (i % 2) == 0:

averted_pre_files_to_combine = [
averted_pre_right_odd_subject,
averted_pre_left_odd_subject,
]
combined_pre_averted_files = mne.concatenate_raws(
averted_pre_files_to_combine
)

# Combine RIGHT => LEFT hand data
combined_pre_averted_files = pd.concat([averted_pre_right_odd_subject, averted_pre_left_odd_subject])

# Create file name for combine files of pre-averted baseline
combined_pre_averted_files_label = (
raw_combined_baseline_data_dir
+ "S"
+ str(i + 1)
+ "-averted_pre_right_left_point_combined_raw.csv"
raw_combined_baseline_data_directory + "S" + str(i + 1) + "-averted_pre_right_left_point_combined_raw.csv"
)

# Save combine pre-averted baseline file to csv
combined_pre_averted_files.to_csv(combined_pre_averted_files_label)


# Check if i (index in this looping) == ODD number that takes EVEN actual subject no, then labeling the file name of hand is LEFT_RIGHT
# For example, i = 1 actually takes S02 and keeps going...

else:

averted_pre_files_to_combine = [
averted_pre_left_odd_subject,
averted_pre_right_odd_subject,
]

combined_pre_averted_files = mne.concatenate_raws(
averted_pre_files_to_combine
)

# Combine LEFT => RIGHT hand data
combined_pre_averted_files = pd.concat([averted_pre_left_odd_subject, averted_pre_right_odd_subject])

# Create file name for combine files of pre-averted baseline
combined_pre_averted_files_label = (
raw_combined_baseline_data_dir
+ "S"
+ str(i + 1)
+ "-averted_pre_left_right_point_combined_raw.csv"
raw_combined_baseline_data_directory + "S" + str(i + 1) + "-averted_pre_left_right_point_combined_raw.csv"
)

combined_pre_averted_files.save( # type: ignore
combined_pre_averted_files_label, overwrite=True
)
# Save combine pre-averted baseline file to csv
combined_pre_averted_files.to_csv(combined_pre_averted_files_label)


# %% [markdown]
# #### Combine post averted baseline

# %%

for i in tqdm(range(16), desc="Combining post averted..."): # type: ignore
# post-averted
for i in tqdm(range(15), desc="Combining post averted..."): # type: ignore
# Pre-averted (for processing subject 1 - 9)
if i < 9:

averted_post_right_odd_subject = mne.io.read_raw_fif(
"EEG-S0" + str(i + 1) + "-averted_post_right_point_raw.fif", verbose=False
)
averted_post_left_odd_subject = mne.io.read_raw_fif(
"EEG-S0" + str(i + 1) + "-averted_post_left_point_raw.fif", verbose=False
)
# Load averted post right
averted_post_right_odd_subject_file_name = "EyeTracker-S0" + str(i + 1) + "-averted_post_right_point_raw.csv"
averted_post_right_odd_subject = pd.read_csv(averted_post_right_odd_subject_file_name)

# Load Load averted post left
averted_post_left_odd_subject_file_name = "EyeTracker-S0" + str(i + 1) + "-averted_post_left_point_raw.csv"
averted_post_left_odd_subject = pd.read_csv(averted_post_left_odd_subject_file_name)

# Check if i (index in this looping) == EVEN number that takes ODD actual subject no, then labeling the file name of hand is RIGHT-LEFT
# For example, i = 0 actually takes S01 and keeps going...
if (i % 2) == 0:

averted_post_files_to_combine = [
averted_post_right_odd_subject,
averted_post_left_odd_subject,
]
combined_post_averted_files = mne.concatenate_raws(
averted_post_files_to_combine
)

# Combine RIGHT => LEFT hand data
combined_post_averted_files = pd.concat([averted_post_right_odd_subject, averted_post_left_odd_subject])

# Create file name for combine files of post-averted baseline
combined_post_averted_files_label = (
"/hpc/igum002/codes/Hyperscanning2-redesign/data/EEG/raw_baseline_data/raw_combined_baseline_data/"
+ "S0"
+ str(i + 1)
+ "-averted_post_right_left_point_combined_raw.fif"
raw_combined_baseline_data_directory + "S0" + str(i + 1) + "-averted_post_right_left_point_combined_raw.csv"
)

# Save combine post-averted baseline file to csv
combined_post_averted_files.to_csv(combined_post_averted_files_label)


# Check if i (index in this looping) == ODD number that takes EVEN actual subject no, then labeling the file name of hand is LEFT_RIGHT
# For example, i = 1 actually takes S02 and keeps going...

else:

averted_post_files_to_combine = [
averted_post_left_odd_subject,
averted_post_right_odd_subject,
]

combined_post_averted_files = mne.concatenate_raws(
averted_post_files_to_combine
)

# Combine LEFT => RIGHT hand data
combined_post_averted_files = pd.concat([averted_post_left_odd_subject, averted_post_right_odd_subject])

# Create file name for combine files of post-averted baseline
combined_post_averted_files_label = (
"/hpc/igum002/codes/Hyperscanning2-redesign/data/EEG/raw_baseline_data/raw_combined_baseline_data/"
+ "S0"
+ str(i + 1)
+ "-averted_post_left_right_point_combined_raw.fif"
raw_combined_baseline_data_directory + "S0" + str(i + 1) + "-averted_post_left_right_point_combined_raw.csv"
)

combined_post_averted_files.save( # type: ignore
combined_post_averted_files_label, overwrite=True
)
# Save combine post-averted baseline file to csv
combined_post_averted_files.to_csv(combined_post_averted_files_label)

# Pre-averted (for processing subject 10 onwards)
else:
# Load averted post right
averted_post_right_odd_subject_file_name = "EyeTracker-S" + str(i + 1) + "-averted_post_right_point_raw.csv"
averted_post_right_odd_subject = pd.read_csv(averted_post_right_odd_subject_file_name)

averted_post_right_odd_subject = mne.io.read_raw_fif(
"EEG-S" + str(i + 1) + "-averted_post_right_point_raw.fif", verbose=False
)
averted_post_left_odd_subject = mne.io.read_raw_fif(
"EEG-S" + str(i + 1) + "-averted_post_left_point_raw.fif", verbose=False
)
averted_post_files_to_combine = [
averted_post_right_odd_subject,
averted_post_left_odd_subject,
]
combined_post_averted_files = mne.concatenate_raws(
averted_post_files_to_combine
)
# Load Load averted post left
averted_post_left_odd_subject_file_name = "EyeTracker-S" + str(i + 1) + "-averted_post_left_point_raw.csv"
averted_post_left_odd_subject = pd.read_csv(averted_post_left_odd_subject_file_name)

# Check if i (index in this looping) == EVEN number that takes ODD actual subject no, then labeling the file name of hand is RIGHT-LEFT
# For example, i = 0 actually takes S01 and keeps going...
if (i % 2) == 0:

averted_post_files_to_combine = [
averted_post_right_odd_subject,
averted_post_left_odd_subject,
]
combined_post_averted_files = mne.concatenate_raws(
averted_post_files_to_combine
)

# Combine RIGHT => LEFT hand data
combined_post_averted_files = pd.concat([averted_post_right_odd_subject, averted_post_left_odd_subject])

# Create file name for combine files of post-averted baseline
combined_post_averted_files_label = (
"/hpc/igum002/codes/Hyperscanning2-redesign/data/EEG/raw_baseline_data/raw_combined_baseline_data/"
+ "S"
+ str(i + 1)
+ "-averted_post_right_left_point_combined_raw.fif"
raw_combined_baseline_data_directory + "S" + str(i + 1) + "-averted_post_right_left_point_combined_raw.csv"
)

# Save combine post-averted baseline file to csv
combined_post_averted_files.to_csv(combined_post_averted_files_label)


# Check if i (index in this looping) == ODD number that takes EVEN actual subject no, then labeling the file name of hand is LEFT_RIGHT
# For example, i = 1 actually takes S02 and keeps going...

else:

averted_post_files_to_combine = [
averted_post_left_odd_subject,
averted_post_right_odd_subject,
]

combined_post_averted_files = mne.concatenate_raws(
averted_post_files_to_combine
)

# Combine LEFT => RIGHT hand data
combined_post_averted_files = pd.concat([averted_post_left_odd_subject, averted_post_right_odd_subject])

# Create file name for combine files of post-averted baseline
combined_post_averted_files_label = (
"/hpc/igum002/codes/Hyperscanning2-redesign/data/EEG/raw_baseline_data/raw_combined_baseline_data/"
+ "S"
+ str(i + 1)
+ "-averted_post_left_right_point_combined_raw.fif"
raw_combined_baseline_data_directory + "S" + str(i + 1) + "-averted_post_left_right_point_combined_raw.csv"
)

combined_post_averted_files.save( # type: ignore
combined_post_averted_files_label, overwrite=True
)
# Save combine post-averted baseline file to csv
combined_post_averted_files.to_csv(combined_post_averted_files_label)


# %% [markdown]
# ### Experimental
Expand Down

0 comments on commit 201ad99

Please sign in to comment.