Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix-remove-spurious-columns #7

Merged
merged 1 commit into from
May 9, 2024
Merged

Fix-remove-spurious-columns #7

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
52 changes: 22 additions & 30 deletions dqc/crossval.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -163,20 +163,18 @@ def _no_calibration(

return preds, pred_probs, label_correctness_scores

def _get_baselines(self, data_with_noisy_labels: pd.DataFrame) -> pd.DataFrame:
def _get_baselines(self, input_data: pd.DataFrame) -> pd.DataFrame:
"""Computes the baseline prediction probabilities using
input label distribution

Args:
data_with_noisy_labels (pd.DataFrame): Input data with
input_data (pd.DataFrame): Input data with
corresponding noisy labels

Returns:
pd.DataFrame: Labels and corresponding probabilities
"""
thresholds_df = (
data_with_noisy_labels[self.y_col_name_int].value_counts(1)
).reset_index()
thresholds_df = (input_data[self.y_col_name_int].value_counts(1)).reset_index()
thresholds_df.columns = [self.y_col_name_int, "probability"]
return thresholds_df

Expand Down Expand Up @@ -270,16 +268,15 @@ def fit_transform(
n_splits = self.n_splits
options["num_samples"] = len(data_with_noisy_labels)

# TBD
# 'max_features': min(options["num_samples"] // 10, options.get("max_features", 1000))

# Make a copy of `data_with_noisy_labels` to prevent accidental modification
input_data = data_with_noisy_labels.copy()
logger.info("Pre-processing the data..")
dp = _DataProcessor(
random_state=self.random_state,
)

data_with_noisy_labels, row_id_col, y_col_name_int = dp._preprocess(
data_with_noisy_labels, y_col_name=y_col_name
input_data, row_id_col, y_col_name_int = dp._preprocess(
input_data, y_col_name=y_col_name
)

# y_col_name_int needs to be accessed downstream
Expand All @@ -294,7 +291,7 @@ def fit_transform(
)

split_indices = _data_splitter(
data_with_noisy_labels,
input_data,
X_col_name=X_col_name,
y_col_name_int=y_col_name_int,
n_splits=self.n_splits,
Expand All @@ -310,18 +307,18 @@ def fit_transform(

if self.calibration_method == "calibrate_using_baseline":
logger.info("Computing baseline predictions for each label..")
baseline_probs = self._get_baselines(data_with_noisy_labels[data_columns])
baseline_probs = self._get_baselines(input_data[data_columns])

# Iterate through kfold splits
for train_index, val_index in tqdm(split_indices):
# Split the data
X_train, X_val = (
data_with_noisy_labels.loc[train_index, X_col_name].values,
data_with_noisy_labels.loc[val_index, X_col_name].values,
input_data.loc[train_index, X_col_name].values,
input_data.loc[val_index, X_col_name].values,
)
y_train, y_val = (
data_with_noisy_labels.loc[train_index, y_col_name_int].values,
data_with_noisy_labels.loc[val_index, y_col_name_int].values,
input_data.loc[train_index, y_col_name_int].values,
input_data.loc[val_index, y_col_name_int].values,
)

# Train the model
Expand All @@ -346,30 +343,25 @@ def fit_transform(
predictions.extend(y_preds)
prediction_probabilities.extend(y_pred_probs)
label_correctness_scores.extend(label_cscores)
row_ids.extend(data_with_noisy_labels.loc[val_index, row_id_col].values)
row_ids.extend(input_data.loc[val_index, row_id_col].values)

# Order dataframe according to `rowids``

row_id_df = pd.DataFrame()
row_id_df[row_id_col] = pd.Series(row_ids)
data_with_noisy_labels = pd.merge(
row_id_df, data_with_noisy_labels, how="left", on=row_id_col
)
input_data = pd.merge(row_id_df, input_data, how="left", on=row_id_col)

# Add results as columns
data_with_noisy_labels["label_correctness_score"] = pd.Series(
label_correctness_scores
)
data_with_noisy_labels["predicted_label"] = pd.Series(predictions)
data_with_noisy_labels["prediction_probability"] = pd.Series(
prediction_probabilities
)
input_data["label_correctness_score"] = pd.Series(label_correctness_scores)
input_data["predicted_label"] = pd.Series(predictions)
input_data["prediction_probability"] = pd.Series(prediction_probabilities)

logger.info("Identifying the correctly labelled samples..")
data_with_noisy_labels["is_label_correct"] = (
data_with_noisy_labels.progress_apply(self._is_confident, axis=1)
input_data["is_label_correct"] = input_data.progress_apply(
self._is_confident, axis=1
)

return dp._postprocess(
data_with_noisy_labels, display_cols=data_columns + self.result_col_list
input_data,
display_cols=list(data_with_noisy_labels.columns) + self.result_col_list,
)
2 changes: 2 additions & 0 deletions tests/test_crossval.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -68,6 +68,8 @@ def run_cvc(
]
)

assert len(data_curated.columns) == len(sampled_data.columns) + 4

# Check data types
assert all(
data_curated[col].dtype == dtype
Expand Down
Loading