diff --git a/19郑静超.xlsx b/19郑静超.xlsx new file mode 100644 index 0000000..4edba59 Binary files /dev/null and b/19郑静超.xlsx differ diff --git a/__pycache__/config.cpython-38.pyc b/__pycache__/config.cpython-38.pyc new file mode 100644 index 0000000..6499df3 Binary files /dev/null and b/__pycache__/config.cpython-38.pyc differ diff --git a/__pycache__/dataset.cpython-38.pyc b/__pycache__/dataset.cpython-38.pyc new file mode 100644 index 0000000..ad1eca7 Binary files /dev/null and b/__pycache__/dataset.cpython-38.pyc differ diff --git a/__pycache__/model_numpy.cpython-38.pyc b/__pycache__/model_numpy.cpython-38.pyc new file mode 100644 index 0000000..f06356e Binary files /dev/null and b/__pycache__/model_numpy.cpython-38.pyc differ diff --git a/__pycache__/train.cpython-38.pyc b/__pycache__/train.cpython-38.pyc new file mode 100644 index 0000000..e374378 Binary files /dev/null and b/__pycache__/train.cpython-38.pyc differ diff --git a/config.py b/config.py index 12a552d..403f252 100644 --- a/config.py +++ b/config.py @@ -13,7 +13,7 @@ MAX_SEQ_LEN = 100 # 句子最大长度（词数） VECTORIZER_TYPE = 'tfidf' # 'tfidf' 或 'bow'（向量化方式） # ==================== 模型相关 ==================== -MODEL_TYPE = 'mlp' # 'mlp' 或 'lr'（模型类型） +MODEL_TYPE = 'lr' # 'mlp' 或 'lr'（模型类型） HIDDEN_SIZE = 64 # MLP隐藏层大小（LR忽略） NUM_CLASSES = 2 # 类别数（正面/负面二分类） KEEP_PROB = 1.0 # Dropout保留概率（LR忽略，设为1即可） @@ -27,11 +27,11 @@ BATCH_SIZE = 64 # 批次大小 USE_CLASS_WEIGHT = True # True=启用类别权重, False=不启用（对比用） # 权重计算公式: n_samples / (n_classes * n_class_i) # 正面评论多所以权重小，负面评论少所以权重大 -CLASS_WEIGHT_POS = 0.73 # 正面类权重（自动计算） -CLASS_WEIGHT_NEG = 1.58 # 负面类权重（自动计算） +CLASS_WEIGHT_POS = 0.83 # 正面类权重（自动计算） +CLASS_WEIGHT_NEG = 1.42 # 负面类权重（自动计算） # ==================== 实验相关 ==================== -RUN_COMPARISON = False # True=运行对比实验, False=运行单个模型 +RUN_COMPARISON = True # True=运行对比实验, False=运行单个模型 COMPARE_MODELS = ['lr', 'mlp'] # 要对比的模型列表 COMPARE_VECTORS = ['bow', 'tfidf'] # 要对比的向量化方式