diff --git a/.idea/runConfigurations/unimore_bda_6.xml b/.idea/runConfigurations/unimore_bda_6.xml
index 580512f..4a7eb2d 100644
--- a/.idea/runConfigurations/unimore_bda_6.xml
+++ b/.idea/runConfigurations/unimore_bda_6.xml
@@ -4,16 +4,9 @@
-
-
-
-
-
-
-
@@ -26,7 +19,7 @@
-
+
diff --git a/unimore_bda_6/__main__.py b/unimore_bda_6/__main__.py
index c59ac1d..60c0ef9 100644
--- a/unimore_bda_6/__main__.py
+++ b/unimore_bda_6/__main__.py
@@ -41,22 +41,22 @@ def main():
slog.debug("Selected sample_func: %s", sample_func.__name__)
for SentimentAnalyzer in [
- # ThreeCheat,
+ ThreeCheat,
+ NLTKSentimentAnalyzer,
TensorflowPolarSentimentAnalyzer,
TensorflowCategorySentimentAnalyzer,
- NLTKSentimentAnalyzer,
]:
slog = logging.getLogger(f"{__name__}.{sample_func.__name__}.{SentimentAnalyzer.__name__}")
slog.debug("Selected SentimentAnalyzer: %s", SentimentAnalyzer.__name__)
for Tokenizer in [
- PottsTokenizer,
PlainTokenizer,
- HuggingBertTokenizer,
- PottsTokenizerWithNegation,
LowercaseTokenizer,
NLTKWordTokenizer,
+ PottsTokenizer,
+ PottsTokenizerWithNegation,
+ HuggingBertTokenizer,
]:
slog = logging.getLogger(f"{__name__}.{sample_func.__name__}.{SentimentAnalyzer.__name__}.{Tokenizer.__name__}")
diff --git a/unimore_bda_6/config.py b/unimore_bda_6/config.py
index bbb5693..419d4e4 100644
--- a/unimore_bda_6/config.py
+++ b/unimore_bda_6/config.py
@@ -34,10 +34,10 @@ def WORKING_SET_SIZE(val: str | None) -> int:
The number of reviews to consider from the database.
Set this to a low number to prevent slowness due to the dataset's huge size.
- Defaults to `10000`.
+ Defaults to `1000000`.
"""
if val is None:
- return 10000
+ return 1000000
try:
return int(val)
except ValueError:
@@ -49,10 +49,10 @@ def TRAINING_SET_SIZE(val: str | None) -> int:
"""
The number of reviews from each category to fetch for the training dataset.
- Defaults to `5000`.
+ Defaults to `4000`.
"""
if val is None:
- return 5000
+ return 4000
try:
return int(val)
except ValueError:
@@ -94,10 +94,10 @@ def TENSORFLOW_MAX_FEATURES(val: str | None) -> int:
"""
The maximum number of features to use in Tensorflow models.
- Defaults to `30000`.
+ Defaults to `300000`.
"""
if val is None:
- return 30000
+ return 300000
try:
return int(val)
except ValueError: