import tensorflow as tfimport pandas as pdimport numpy as npimport matplotlib as pltfrom sklearn.model_selection import train_test_splitfrom tensorflow import kerasfrom tensorflow.keras.models import Sequentialfrom tensorflow.keras.layers import Densefrom sklearn.preprocessing import StandardScalerimport functoolsLABEL_COLUMN = 'Endstage'LABELS = [1, 2, 3, 4]x = pd.read_csv('HCVnew.csv', index_col=False)def get_dataset(file_path, **kwargs): dataset = tf.data.experimental.make_csv_dataset( file_path, batch_size=35, # Artificially small to make examples easier to show. label_name=LABEL_COLUMN, na_value="?", num_epochs=1, ignore_errors=True, **kwargs) return datasetSELECT_COLUMNS = ["Alter", "Gender", "BMI", "Fever", "Nausea", "Fatigue", "WBC", "RBC", "HGB", "Plat", "AST1", "ALT1", "ALT4", "ALT12", "ALT24", "ALT36", "ALT48", "ALT24w", "RNABase", "RNA4", "Baseline", "Endstage"]DEFAULTS = [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0]temp_dataset = get_dataset("HCVnew.csv", select_columns=SELECT_COLUMNS, column_defaults=DEFAULTS)def pack(features, label): return tf.stack(list(features.values()), axis=-1), labelpacked_dataset = temp_dataset.map(pack)"""for features, labels in packed_dataset.take(1): print(features.numpy()) print() print(labels.numpy())"""NUMERIC_FEATURES = ["Alter", "Gender","BMI", "Fever", "Nausea", "Fatigue", "WBC", "RBC", "HGB", "Plat", "AST1", "ALT1", "ALT4", "ALT12", "ALT24", "ALT36", "ALT48", "ALT24w", "RNABase", "RNA4", "Baseline", "Endstage"]您好，我正在嘗試構(gòu)建一個(gè)神經(jīng)網(wǎng)絡(luò)，該神經(jīng)網(wǎng)絡(luò)可以根據(jù)包含患者信息的csv文件預(yù)測(cè)丙型肝炎，我無(wú)法修復(fù)錯(cuò)誤...我收到錯(cuò)誤：KeyError'Endstage'，而Endstage是包含相應(yīng)值（介于1和4之間）并用作標(biāo)簽列的csv列。如果有人有一個(gè)想法可以解決我的問(wèn)題，那么請(qǐng)告訴我。非常感謝您的幫助！
1 回答
小怪獸愛(ài)吃肉
TA貢獻(xiàn)1852條經(jīng)驗(yàn) 獲得超1個(gè)贊
這是因?yàn)槭悄愕臉?biāo)簽列，框架通過(guò)從數(shù)據(jù)集中刪除（彈出）它來(lái)幫你一個(gè)忙。否則，您的訓(xùn)練數(shù)據(jù)集也將具有目標(biāo)類(lèi)，使其無(wú)用。Endstage
將其從任何其他位置刪除，使其進(jìn)入訓(xùn)練集功能。NUMERIC_FEATURES
[編輯]
OP在后續(xù)問(wèn)題（在評(píng)論中）中詢(xún)問(wèn)了為什么在修復(fù)初始問(wèn)題后，他會(huì)遇到錯(cuò)誤：
值錯(cuò)誤：特征數(shù)字不在特征字典中
從外觀上看，調(diào)用的特征是通過(guò)調(diào)用 來(lái)生成的。后者用于創(chuàng)建 和 ，但這些從不使用。然而，這句話(huà)：numericPackNumericFeaturespacked_train_datapacked_test_data
numeric_column = tf.feature_column.numeric_column('numeric', normalizer_fn=normalizer, shape=[len(NUMERIC_FEATURES)])假設(shè)數(shù)據(jù)在那里 - 因此出現(xiàn)錯(cuò)誤。
添加回答
舉報(bào)
0/150
提交
取消