使用 TensorFlow¶
关于本教程¶
本教程展示了如何将 YDF 决策森林模型与 TensorFlow 神经网络模型结合起来。这种组合模型通常用于提高模型质量并处理非结构化数据。
首先，我们展示了如何在预训练的神经网络模型上“堆叠”一个决策森林模型。这种方法对于“预训练嵌入”方法特别有用：我们不是直接在特征上训练模型，而是训练或使用一个已经在大量相似数据上训练的现有模型。第一个模型的输出用作第二个模型的输入，后者是在我们有限的数据集上训练的。“预训练嵌入”方法也简化了文本或图像的模型开发。虽然将文本或图像输入模型需要一些努力，但嵌入（也称为中间表示）则更容易进行训练。
在第二部分中，我们展示了如何集成多个决策森林和神经网络模型。使用不同模型类型的集成可以利用它们的互补优势，因为决策森林在某些示例中表现出色，而神经网络在其他情况下更具优势。
设置¶
In [ ]:
Copied!
# 安装YDF和TensorFlow
!pip install ydf -U -q
!pip install tensorflow -U -q
# 安装YDF和TensorFlow
!pip install ydf -U -q
!pip install tensorflow -U -q
堆叠决策森林和预训练神经网络模型¶
在这个例子中，我们在 斯坦福情感树库 文本分类数据集上训练模型。 预测任务是将电影评论分类为正面（标签 1）或负面（标签 0）。 数据集可以在 TensorFlow 数据集 中找到。
In [ ]:
Copied!
# 安装 TensorFlow 数据集
!pip install tensorflow-datasets -U -q
# 安装 TensorFlow 数据集
!pip install tensorflow-datasets -U -q
In [2]:
Copied!
# 加载斯坦福情感树库数据集
import tensorflow_datasets as tfds
raw_dataset = tfds.load("glue/sst2")
raw_train_dataset = raw_dataset["train"].batch(200)
raw_test_dataset = raw_dataset["validation"].batch(200)
# Note: The `raw_dataset["test"]` fold does not have labels. Therefore, we use the
# "validation" fold for testing.
# 显示前3个测试示例。
for example in raw_test_dataset.rebatch(1).take(3):
print("label:", example["label"].numpy())
print("sentence:", example["sentence"].numpy())
print("")
# 加载斯坦福情感树库数据集
import tensorflow_datasets as tfds
raw_dataset = tfds.load("glue/sst2")
raw_train_dataset = raw_dataset["train"].batch(200)
raw_test_dataset = raw_dataset["validation"].batch(200)
# Note: The `raw_dataset["test"]` fold does not have labels. Therefore, we use the
# "validation" fold for testing.
# 显示前3个测试示例。
for example in raw_test_dataset.rebatch(1).take(3):
print("label:", example["label"].numpy())
print("sentence:", example["sentence"].numpy())
print("")
label: [0] sentence: [b'a valueless kiddie paean to pro basketball underwritten by the nba . '] label: [1] sentence: [b"featuring a dangerously seductive performance from the great daniel auteuil , `` sade '' covers the same period as kaufmann 's `` quills '' with more unsettlingly realistic results . "] label: [0] sentence: [b'i am sorry that i was unable to get the full brunt of the comedy . ']
2024-04-12 10:19:56.151619: W tensorflow/core/kernels/data/cache_dataset_ops.cc:858] The calling iterator did not fully read the dataset being cached. In order to avoid unexpected truncation of the dataset, the partially cached contents of the dataset will be discarded. This can happen if you have an input pipeline similar to `dataset.cache().take(k).repeat()`. You should use `dataset.take(k).cache().repeat()` instead. 2024-04-12 10:19:56.151977: W tensorflow/core/framework/local_rendezvous.cc:404] Local rendezvous is aborting with status: OUT_OF_RANGE: End of sequence
斯坦福情感树库数据集相对较小。因此，我们将不直接在其上进行训练，而是使用一个已经训练好的文本模型。例如，我们将使用通用句子编码器神经网络，它是在一个更大数据集上训练的。通过重用USE，我们可以快速训练文本模型而无需大型数据集。
我们使用在TensorFlow Hub网站上提供的通用句子编码器。对于其他类型的非结构化数据（例如，图像或音频），相应的预训练模型可以在TensorFlow Hub网站上找到。
In [ ]:
Copied!
# 安装 TensorFlow Hub
!pip install tensorflow-hub -U -q
# 安装 TensorFlow Hub
!pip install tensorflow-hub -U -q
我们加载通用句子编码器。
In [3]:
Copied!
import tensorflow_hub as hub
pretrained_neural_network_model = hub.load("https://tfhub.dev/google/universal-sentence-encoder/4")
import tensorflow_hub as hub
pretrained_neural_network_model = hub.load("https://tfhub.dev/google/universal-sentence-encoder/4")
我们应用通用句子编码器将原始文本转换为嵌入表示。
In [4]:
Copied!
import numpy as np
def apply_neural_network_model(data):
return {
"embedding": pretrained_neural_network_model(data["sentence"]),
"label": data["label"],
}
processed_train_dataset = raw_train_dataset.map(apply_neural_network_model)
processed_test_dataset = raw_test_dataset.map(apply_neural_network_model)
# 显示前3个已处理的测试示例。
for example in processed_test_dataset.rebatch(1).take(3):
print("label:", example["label"].numpy())
print("embedding:", np.array2string(example["embedding"].numpy(), threshold = 10))
print("")
import numpy as np
def apply_neural_network_model(data):
return {
"embedding": pretrained_neural_network_model(data["sentence"]),
"label": data["label"],
}
processed_train_dataset = raw_train_dataset.map(apply_neural_network_model)
processed_test_dataset = raw_test_dataset.map(apply_neural_network_model)
# 显示前3个已处理的测试示例。
for example in processed_test_dataset.rebatch(1).take(3):
print("label:", example["label"].numpy())
print("embedding:", np.array2string(example["embedding"].numpy(), threshold = 10))
print("")
label: [0] embedding: [[ 0.0325069 0.02350717 0.0928923 ... -0.05117338 -0.0959709 0.01819227]] label: [1] embedding: [[-0.02955496 0.04451125 0.03023855 ... 0.02529242 0.06243655 0.0004012 ]] label: [0] embedding: [[-0.02897574 -0.0021273 0.02524822 ... -0.00667133 0.07221754 0.05386261]]
2024-04-12 10:20:03.889224: W tensorflow/core/kernels/data/cache_dataset_ops.cc:858] The calling iterator did not fully read the dataset being cached. In order to avoid unexpected truncation of the dataset, the partially cached contents of the dataset will be discarded. This can happen if you have an input pipeline similar to `dataset.cache().take(k).repeat()`. You should use `dataset.take(k).cache().repeat()` instead. 2024-04-12 10:20:03.889676: W tensorflow/core/framework/local_rendezvous.cc:404] Local rendezvous is aborting with status: OUT_OF_RANGE: End of sequence
然后，我们在神经网络的输出上训练一个梯度提升树模型。
In [5]:
Copied!
import ydf
decision_forest_model = ydf.GradientBoostedTreesLearner(label="label", num_trees=100).train(processed_train_dataset)
import ydf
decision_forest_model = ydf.GradientBoostedTreesLearner(label="label", num_trees=100).train(processed_train_dataset)
Train model on 67349 examples Model trained in 0:00:54.704927
注意到YDF模型是直接在TensorFlow数据集上进行训练的。
仔细观察更为重要。
In [6]:
Copied!
decision_forest_model.describe()
decision_forest_model.describe()
Out[6]:
Name : GRADIENT_BOOSTED_TREES
Task : CLASSIFICATION
Label : label
Features (512) : embedding.000_of_512 embedding.001_of_512 embedding.002_of_512 embedding.003_of_512 embedding.004_of_512 embedding.005_of_512 embedding.006_of_512 embedding.007_of_512 embedding.008_of_512 embedding.009_of_512 embedding.010_of_512 embedding.011_of_512 embedding.012_of_512 embedding.013_of_512 embedding.014_of_512 embedding.015_of_512 embedding.016_of_512 embedding.017_of_512 embedding.018_of_512 embedding.019_of_512 embedding.020_of_512 embedding.021_of_512 embedding.022_of_512 embedding.023_of_512 embedding.024_of_512 embedding.025_of_512 embedding.026_of_512 embedding.027_of_512 embedding.028_of_512 embedding.029_of_512 embedding.030_of_512 embedding.031_of_512 embedding.032_of_512 embedding.033_of_512 embedding.034_of_512 embedding.035_of_512 embedding.036_of_512 embedding.037_of_512 embedding.038_of_512 embedding.039_of_512 embedding.040_of_512 embedding.041_of_512 embedding.042_of_512 embedding.043_of_512 embedding.044_of_512 embedding.045_of_512 embedding.046_of_512 embedding.047_of_512 embedding.048_of_512 embedding.049_of_512 embedding.050_of_512 embedding.051_of_512 embedding.052_of_512 embedding.053_of_512 embedding.054_of_512 embedding.055_of_512 embedding.056_of_512 embedding.057_of_512 embedding.058_of_512 embedding.059_of_512 embedding.060_of_512 embedding.061_of_512 embedding.062_of_512 embedding.063_of_512 embedding.064_of_512 embedding.065_of_512 embedding.066_of_512 embedding.067_of_512 embedding.068_of_512 embedding.069_of_512 embedding.070_of_512 embedding.071_of_512 embedding.072_of_512 embedding.073_of_512 embedding.074_of_512 embedding.075_of_512 embedding.076_of_512 embedding.077_of_512 embedding.078_of_512 embedding.079_of_512 embedding.080_of_512 embedding.081_of_512 embedding.082_of_512 embedding.083_of_512 embedding.084_of_512 embedding.085_of_512 embedding.086_of_512 embedding.087_of_512 embedding.088_of_512 embedding.089_of_512 embedding.090_of_512 embedding.091_of_512 embedding.092_of_512 embedding.093_of_512 embedding.094_of_512 embedding.095_of_512 embedding.096_of_512 embedding.097_of_512 embedding.098_of_512 embedding.099_of_512 embedding.100_of_512 embedding.101_of_512 embedding.102_of_512 embedding.103_of_512 embedding.104_of_512 embedding.105_of_512 embedding.106_of_512 embedding.107_of_512 embedding.108_of_512 embedding.109_of_512 embedding.110_of_512 embedding.111_of_512 embedding.112_of_512 embedding.113_of_512 embedding.114_of_512 embedding.115_of_512 embedding.116_of_512 embedding.117_of_512 embedding.118_of_512 embedding.119_of_512 embedding.120_of_512 embedding.121_of_512 embedding.122_of_512 embedding.123_of_512 embedding.124_of_512 embedding.125_of_512 embedding.126_of_512 embedding.127_of_512 embedding.128_of_512 embedding.129_of_512 embedding.130_of_512 embedding.131_of_512 embedding.132_of_512 embedding.133_of_512 embedding.134_of_512 embedding.135_of_512 embedding.136_of_512 embedding.137_of_512 embedding.138_of_512 embedding.139_of_512 embedding.140_of_512 embedding.141_of_512 embedding.142_of_512 embedding.143_of_512 embedding.144_of_512 embedding.145_of_512 embedding.146_of_512 embedding.147_of_512 embedding.148_of_512 embedding.149_of_512 embedding.150_of_512 embedding.151_of_512 embedding.152_of_512 embedding.153_of_512 embedding.154_of_512 embedding.155_of_512 embedding.156_of_512 embedding.157_of_512 embedding.158_of_512 embedding.159_of_512 embedding.160_of_512 embedding.161_of_512 embedding.162_of_512 embedding.163_of_512 embedding.164_of_512 embedding.165_of_512 embedding.166_of_512 embedding.167_of_512 embedding.168_of_512 embedding.169_of_512 embedding.170_of_512 embedding.171_of_512 embedding.172_of_512 embedding.173_of_512 embedding.174_of_512 embedding.175_of_512 embedding.176_of_512 embedding.177_of_512 embedding.178_of_512 embedding.179_of_512 embedding.180_of_512 embedding.181_of_512 embedding.182_of_512 embedding.183_of_512 embedding.184_of_512 embedding.185_of_512 embedding.186_of_512 embedding.187_of_512 embedding.188_of_512 embedding.189_of_512 embedding.190_of_512 embedding.191_of_512 embedding.192_of_512 embedding.193_of_512 embedding.194_of_512 embedding.195_of_512 embedding.196_of_512 embedding.197_of_512 embedding.198_of_512 embedding.199_of_512 embedding.200_of_512 embedding.201_of_512 embedding.202_of_512 embedding.203_of_512 embedding.204_of_512 embedding.205_of_512 embedding.206_of_512 embedding.207_of_512 embedding.208_of_512 embedding.209_of_512 embedding.210_of_512 embedding.211_of_512 embedding.212_of_512 embedding.213_of_512 embedding.214_of_512 embedding.215_of_512 embedding.216_of_512 embedding.217_of_512 embedding.218_of_512 embedding.219_of_512 embedding.220_of_512 embedding.221_of_512 embedding.222_of_512 embedding.223_of_512 embedding.224_of_512 embedding.225_of_512 embedding.226_of_512 embedding.227_of_512 embedding.228_of_512 embedding.229_of_512 embedding.230_of_512 embedding.231_of_512 embedding.232_of_512 embedding.233_of_512 embedding.234_of_512 embedding.235_of_512 embedding.236_of_512 embedding.237_of_512 embedding.238_of_512 embedding.239_of_512 embedding.240_of_512 embedding.241_of_512 embedding.242_of_512 embedding.243_of_512 embedding.244_of_512 embedding.245_of_512 embedding.246_of_512 embedding.247_of_512 embedding.248_of_512 embedding.249_of_512 embedding.250_of_512 embedding.251_of_512 embedding.252_of_512 embedding.253_of_512 embedding.254_of_512 embedding.255_of_512 embedding.256_of_512 embedding.257_of_512 embedding.258_of_512 embedding.259_of_512 embedding.260_of_512 embedding.261_of_512 embedding.262_of_512 embedding.263_of_512 embedding.264_of_512 embedding.265_of_512 embedding.266_of_512 embedding.267_of_512 embedding.268_of_512 embedding.269_of_512 embedding.270_of_512 embedding.271_of_512 embedding.272_of_512 embedding.273_of_512 embedding.274_of_512 embedding.275_of_512 embedding.276_of_512 embedding.277_of_512 embedding.278_of_512 embedding.279_of_512 embedding.280_of_512 embedding.281_of_512 embedding.282_of_512 embedding.283_of_512 embedding.284_of_512 embedding.285_of_512 embedding.286_of_512 embedding.287_of_512 embedding.288_of_512 embedding.289_of_512 embedding.290_of_512 embedding.291_of_512 embedding.292_of_512 embedding.293_of_512 embedding.294_of_512 embedding.295_of_512 embedding.296_of_512 embedding.297_of_512 embedding.298_of_512 embedding.299_of_512 embedding.300_of_512 embedding.301_of_512 embedding.302_of_512 embedding.303_of_512 embedding.304_of_512 embedding.305_of_512 embedding.306_of_512 embedding.307_of_512 embedding.308_of_512 embedding.309_of_512 embedding.310_of_512 embedding.311_of_512 embedding.312_of_512 embedding.313_of_512 embedding.314_of_512 embedding.315_of_512 embedding.316_of_512 embedding.317_of_512 embedding.318_of_512 embedding.319_of_512 embedding.320_of_512 embedding.321_of_512 embedding.322_of_512 embedding.323_of_512 embedding.324_of_512 embedding.325_of_512 embedding.326_of_512 embedding.327_of_512 embedding.328_of_512 embedding.329_of_512 embedding.330_of_512 embedding.331_of_512 embedding.332_of_512 embedding.333_of_512 embedding.334_of_512 embedding.335_of_512 embedding.336_of_512 embedding.337_of_512 embedding.338_of_512 embedding.339_of_512 embedding.340_of_512 embedding.341_of_512 embedding.342_of_512 embedding.343_of_512 embedding.344_of_512 embedding.345_of_512 embedding.346_of_512 embedding.347_of_512 embedding.348_of_512 embedding.349_of_512 embedding.350_of_512 embedding.351_of_512 embedding.352_of_512 embedding.353_of_512 embedding.354_of_512 embedding.355_of_512 embedding.356_of_512 embedding.357_of_512 embedding.358_of_512 embedding.359_of_512 embedding.360_of_512 embedding.361_of_512 embedding.362_of_512 embedding.363_of_512 embedding.364_of_512 embedding.365_of_512 embedding.366_of_512 embedding.367_of_512 embedding.368_of_512 embedding.369_of_512 embedding.370_of_512 embedding.371_of_512 embedding.372_of_512 embedding.373_of_512 embedding.374_of_512 embedding.375_of_512 embedding.376_of_512 embedding.377_of_512 embedding.378_of_512 embedding.379_of_512 embedding.380_of_512 embedding.381_of_512 embedding.382_of_512 embedding.383_of_512 embedding.384_of_512 embedding.385_of_512 embedding.386_of_512 embedding.387_of_512 embedding.388_of_512 embedding.389_of_512 embedding.390_of_512 embedding.391_of_512 embedding.392_of_512 embedding.393_of_512 embedding.394_of_512 embedding.395_of_512 embedding.396_of_512 embedding.397_of_512 embedding.398_of_512 embedding.399_of_512 embedding.400_of_512 embedding.401_of_512 embedding.402_of_512 embedding.403_of_512 embedding.404_of_512 embedding.405_of_512 embedding.406_of_512 embedding.407_of_512 embedding.408_of_512 embedding.409_of_512 embedding.410_of_512 embedding.411_of_512 embedding.412_of_512 embedding.413_of_512 embedding.414_of_512 embedding.415_of_512 embedding.416_of_512 embedding.417_of_512 embedding.418_of_512 embedding.419_of_512 embedding.420_of_512 embedding.421_of_512 embedding.422_of_512 embedding.423_of_512 embedding.424_of_512 embedding.425_of_512 embedding.426_of_512 embedding.427_of_512 embedding.428_of_512 embedding.429_of_512 embedding.430_of_512 embedding.431_of_512 embedding.432_of_512 embedding.433_of_512 embedding.434_of_512 embedding.435_of_512 embedding.436_of_512 embedding.437_of_512 embedding.438_of_512 embedding.439_of_512 embedding.440_of_512 embedding.441_of_512 embedding.442_of_512 embedding.443_of_512 embedding.444_of_512 embedding.445_of_512 embedding.446_of_512 embedding.447_of_512 embedding.448_of_512 embedding.449_of_512 embedding.450_of_512 embedding.451_of_512 embedding.452_of_512 embedding.453_of_512 embedding.454_of_512 embedding.455_of_512 embedding.456_of_512 embedding.457_of_512 embedding.458_of_512 embedding.459_of_512 embedding.460_of_512 embedding.461_of_512 embedding.462_of_512 embedding.463_of_512 embedding.464_of_512 embedding.465_of_512 embedding.466_of_512 embedding.467_of_512 embedding.468_of_512 embedding.469_of_512 embedding.470_of_512 embedding.471_of_512 embedding.472_of_512 embedding.473_of_512 embedding.474_of_512 embedding.475_of_512 embedding.476_of_512 embedding.477_of_512 embedding.478_of_512 embedding.479_of_512 embedding.480_of_512 embedding.481_of_512 embedding.482_of_512 embedding.483_of_512 embedding.484_of_512 embedding.485_of_512 embedding.486_of_512 embedding.487_of_512 embedding.488_of_512 embedding.489_of_512 embedding.490_of_512 embedding.491_of_512 embedding.492_of_512 embedding.493_of_512 embedding.494_of_512 embedding.495_of_512 embedding.496_of_512 embedding.497_of_512 embedding.498_of_512 embedding.499_of_512 embedding.500_of_512 embedding.501_of_512 embedding.502_of_512 embedding.503_of_512 embedding.504_of_512 embedding.505_of_512 embedding.506_of_512 embedding.507_of_512 embedding.508_of_512 embedding.509_of_512 embedding.510_of_512 embedding.511_of_512
Weights : None
Trained with tuner : No
Model size : 1643 kB
Task : CLASSIFICATION
Label : label
Features (512) : embedding.000_of_512 embedding.001_of_512 embedding.002_of_512 embedding.003_of_512 embedding.004_of_512 embedding.005_of_512 embedding.006_of_512 embedding.007_of_512 embedding.008_of_512 embedding.009_of_512 embedding.010_of_512 embedding.011_of_512 embedding.012_of_512 embedding.013_of_512 embedding.014_of_512 embedding.015_of_512 embedding.016_of_512 embedding.017_of_512 embedding.018_of_512 embedding.019_of_512 embedding.020_of_512 embedding.021_of_512 embedding.022_of_512 embedding.023_of_512 embedding.024_of_512 embedding.025_of_512 embedding.026_of_512 embedding.027_of_512 embedding.028_of_512 embedding.029_of_512 embedding.030_of_512 embedding.031_of_512 embedding.032_of_512 embedding.033_of_512 embedding.034_of_512 embedding.035_of_512 embedding.036_of_512 embedding.037_of_512 embedding.038_of_512 embedding.039_of_512 embedding.040_of_512 embedding.041_of_512 embedding.042_of_512 embedding.043_of_512 embedding.044_of_512 embedding.045_of_512 embedding.046_of_512 embedding.047_of_512 embedding.048_of_512 embedding.049_of_512 embedding.050_of_512 embedding.051_of_512 embedding.052_of_512 embedding.053_of_512 embedding.054_of_512 embedding.055_of_512 embedding.056_of_512 embedding.057_of_512 embedding.058_of_512 embedding.059_of_512 embedding.060_of_512 embedding.061_of_512 embedding.062_of_512 embedding.063_of_512 embedding.064_of_512 embedding.065_of_512 embedding.066_of_512 embedding.067_of_512 embedding.068_of_512 embedding.069_of_512 embedding.070_of_512 embedding.071_of_512 embedding.072_of_512 embedding.073_of_512 embedding.074_of_512 embedding.075_of_512 embedding.076_of_512 embedding.077_of_512 embedding.078_of_512 embedding.079_of_512 embedding.080_of_512 embedding.081_of_512 embedding.082_of_512 embedding.083_of_512 embedding.084_of_512 embedding.085_of_512 embedding.086_of_512 embedding.087_of_512 embedding.088_of_512 embedding.089_of_512 embedding.090_of_512 embedding.091_of_512 embedding.092_of_512 embedding.093_of_512 embedding.094_of_512 embedding.095_of_512 embedding.096_of_512 embedding.097_of_512 embedding.098_of_512 embedding.099_of_512 embedding.100_of_512 embedding.101_of_512 embedding.102_of_512 embedding.103_of_512 embedding.104_of_512 embedding.105_of_512 embedding.106_of_512 embedding.107_of_512 embedding.108_of_512 embedding.109_of_512 embedding.110_of_512 embedding.111_of_512 embedding.112_of_512 embedding.113_of_512 embedding.114_of_512 embedding.115_of_512 embedding.116_of_512 embedding.117_of_512 embedding.118_of_512 embedding.119_of_512 embedding.120_of_512 embedding.121_of_512 embedding.122_of_512 embedding.123_of_512 embedding.124_of_512 embedding.125_of_512 embedding.126_of_512 embedding.127_of_512 embedding.128_of_512 embedding.129_of_512 embedding.130_of_512 embedding.131_of_512 embedding.132_of_512 embedding.133_of_512 embedding.134_of_512 embedding.135_of_512 embedding.136_of_512 embedding.137_of_512 embedding.138_of_512 embedding.139_of_512 embedding.140_of_512 embedding.141_of_512 embedding.142_of_512 embedding.143_of_512 embedding.144_of_512 embedding.145_of_512 embedding.146_of_512 embedding.147_of_512 embedding.148_of_512 embedding.149_of_512 embedding.150_of_512 embedding.151_of_512 embedding.152_of_512 embedding.153_of_512 embedding.154_of_512 embedding.155_of_512 embedding.156_of_512 embedding.157_of_512 embedding.158_of_512 embedding.159_of_512 embedding.160_of_512 embedding.161_of_512 embedding.162_of_512 embedding.163_of_512 embedding.164_of_512 embedding.165_of_512 embedding.166_of_512 embedding.167_of_512 embedding.168_of_512 embedding.169_of_512 embedding.170_of_512 embedding.171_of_512 embedding.172_of_512 embedding.173_of_512 embedding.174_of_512 embedding.175_of_512 embedding.176_of_512 embedding.177_of_512 embedding.178_of_512 embedding.179_of_512 embedding.180_of_512 embedding.181_of_512 embedding.182_of_512 embedding.183_of_512 embedding.184_of_512 embedding.185_of_512 embedding.186_of_512 embedding.187_of_512 embedding.188_of_512 embedding.189_of_512 embedding.190_of_512 embedding.191_of_512 embedding.192_of_512 embedding.193_of_512 embedding.194_of_512 embedding.195_of_512 embedding.196_of_512 embedding.197_of_512 embedding.198_of_512 embedding.199_of_512 embedding.200_of_512 embedding.201_of_512 embedding.202_of_512 embedding.203_of_512 embedding.204_of_512 embedding.205_of_512 embedding.206_of_512 embedding.207_of_512 embedding.208_of_512 embedding.209_of_512 embedding.210_of_512 embedding.211_of_512 embedding.212_of_512 embedding.213_of_512 embedding.214_of_512 embedding.215_of_512 embedding.216_of_512 embedding.217_of_512 embedding.218_of_512 embedding.219_of_512 embedding.220_of_512 embedding.221_of_512 embedding.222_of_512 embedding.223_of_512 embedding.224_of_512 embedding.225_of_512 embedding.226_of_512 embedding.227_of_512 embedding.228_of_512 embedding.229_of_512 embedding.230_of_512 embedding.231_of_512 embedding.232_of_512 embedding.233_of_512 embedding.234_of_512 embedding.235_of_512 embedding.236_of_512 embedding.237_of_512 embedding.238_of_512 embedding.239_of_512 embedding.240_of_512 embedding.241_of_512 embedding.242_of_512 embedding.243_of_512 embedding.244_of_512 embedding.245_of_512 embedding.246_of_512 embedding.247_of_512 embedding.248_of_512 embedding.249_of_512 embedding.250_of_512 embedding.251_of_512 embedding.252_of_512 embedding.253_of_512 embedding.254_of_512 embedding.255_of_512 embedding.256_of_512 embedding.257_of_512 embedding.258_of_512 embedding.259_of_512 embedding.260_of_512 embedding.261_of_512 embedding.262_of_512 embedding.263_of_512 embedding.264_of_512 embedding.265_of_512 embedding.266_of_512 embedding.267_of_512 embedding.268_of_512 embedding.269_of_512 embedding.270_of_512 embedding.271_of_512 embedding.272_of_512 embedding.273_of_512 embedding.274_of_512 embedding.275_of_512 embedding.276_of_512 embedding.277_of_512 embedding.278_of_512 embedding.279_of_512 embedding.280_of_512 embedding.281_of_512 embedding.282_of_512 embedding.283_of_512 embedding.284_of_512 embedding.285_of_512 embedding.286_of_512 embedding.287_of_512 embedding.288_of_512 embedding.289_of_512 embedding.290_of_512 embedding.291_of_512 embedding.292_of_512 embedding.293_of_512 embedding.294_of_512 embedding.295_of_512 embedding.296_of_512 embedding.297_of_512 embedding.298_of_512 embedding.299_of_512 embedding.300_of_512 embedding.301_of_512 embedding.302_of_512 embedding.303_of_512 embedding.304_of_512 embedding.305_of_512 embedding.306_of_512 embedding.307_of_512 embedding.308_of_512 embedding.309_of_512 embedding.310_of_512 embedding.311_of_512 embedding.312_of_512 embedding.313_of_512 embedding.314_of_512 embedding.315_of_512 embedding.316_of_512 embedding.317_of_512 embedding.318_of_512 embedding.319_of_512 embedding.320_of_512 embedding.321_of_512 embedding.322_of_512 embedding.323_of_512 embedding.324_of_512 embedding.325_of_512 embedding.326_of_512 embedding.327_of_512 embedding.328_of_512 embedding.329_of_512 embedding.330_of_512 embedding.331_of_512 embedding.332_of_512 embedding.333_of_512 embedding.334_of_512 embedding.335_of_512 embedding.336_of_512 embedding.337_of_512 embedding.338_of_512 embedding.339_of_512 embedding.340_of_512 embedding.341_of_512 embedding.342_of_512 embedding.343_of_512 embedding.344_of_512 embedding.345_of_512 embedding.346_of_512 embedding.347_of_512 embedding.348_of_512 embedding.349_of_512 embedding.350_of_512 embedding.351_of_512 embedding.352_of_512 embedding.353_of_512 embedding.354_of_512 embedding.355_of_512 embedding.356_of_512 embedding.357_of_512 embedding.358_of_512 embedding.359_of_512 embedding.360_of_512 embedding.361_of_512 embedding.362_of_512 embedding.363_of_512 embedding.364_of_512 embedding.365_of_512 embedding.366_of_512 embedding.367_of_512 embedding.368_of_512 embedding.369_of_512 embedding.370_of_512 embedding.371_of_512 embedding.372_of_512 embedding.373_of_512 embedding.374_of_512 embedding.375_of_512 embedding.376_of_512 embedding.377_of_512 embedding.378_of_512 embedding.379_of_512 embedding.380_of_512 embedding.381_of_512 embedding.382_of_512 embedding.383_of_512 embedding.384_of_512 embedding.385_of_512 embedding.386_of_512 embedding.387_of_512 embedding.388_of_512 embedding.389_of_512 embedding.390_of_512 embedding.391_of_512 embedding.392_of_512 embedding.393_of_512 embedding.394_of_512 embedding.395_of_512 embedding.396_of_512 embedding.397_of_512 embedding.398_of_512 embedding.399_of_512 embedding.400_of_512 embedding.401_of_512 embedding.402_of_512 embedding.403_of_512 embedding.404_of_512 embedding.405_of_512 embedding.406_of_512 embedding.407_of_512 embedding.408_of_512 embedding.409_of_512 embedding.410_of_512 embedding.411_of_512 embedding.412_of_512 embedding.413_of_512 embedding.414_of_512 embedding.415_of_512 embedding.416_of_512 embedding.417_of_512 embedding.418_of_512 embedding.419_of_512 embedding.420_of_512 embedding.421_of_512 embedding.422_of_512 embedding.423_of_512 embedding.424_of_512 embedding.425_of_512 embedding.426_of_512 embedding.427_of_512 embedding.428_of_512 embedding.429_of_512 embedding.430_of_512 embedding.431_of_512 embedding.432_of_512 embedding.433_of_512 embedding.434_of_512 embedding.435_of_512 embedding.436_of_512 embedding.437_of_512 embedding.438_of_512 embedding.439_of_512 embedding.440_of_512 embedding.441_of_512 embedding.442_of_512 embedding.443_of_512 embedding.444_of_512 embedding.445_of_512 embedding.446_of_512 embedding.447_of_512 embedding.448_of_512 embedding.449_of_512 embedding.450_of_512 embedding.451_of_512 embedding.452_of_512 embedding.453_of_512 embedding.454_of_512 embedding.455_of_512 embedding.456_of_512 embedding.457_of_512 embedding.458_of_512 embedding.459_of_512 embedding.460_of_512 embedding.461_of_512 embedding.462_of_512 embedding.463_of_512 embedding.464_of_512 embedding.465_of_512 embedding.466_of_512 embedding.467_of_512 embedding.468_of_512 embedding.469_of_512 embedding.470_of_512 embedding.471_of_512 embedding.472_of_512 embedding.473_of_512 embedding.474_of_512 embedding.475_of_512 embedding.476_of_512 embedding.477_of_512 embedding.478_of_512 embedding.479_of_512 embedding.480_of_512 embedding.481_of_512 embedding.482_of_512 embedding.483_of_512 embedding.484_of_512 embedding.485_of_512 embedding.486_of_512 embedding.487_of_512 embedding.488_of_512 embedding.489_of_512 embedding.490_of_512 embedding.491_of_512 embedding.492_of_512 embedding.493_of_512 embedding.494_of_512 embedding.495_of_512 embedding.496_of_512 embedding.497_of_512 embedding.498_of_512 embedding.499_of_512 embedding.500_of_512 embedding.501_of_512 embedding.502_of_512 embedding.503_of_512 embedding.504_of_512 embedding.505_of_512 embedding.506_of_512 embedding.507_of_512 embedding.508_of_512 embedding.509_of_512 embedding.510_of_512 embedding.511_of_512
Weights : None
Trained with tuner : No
Model size : 1643 kB
Number of records: 67349 Number of columns: 513 Number of columns by type: NUMERICAL: 512 (99.8051%) CATEGORICAL: 1 (0.194932%) Columns: NUMERICAL: 512 (99.8051%) 1: "embedding.000_of_512" NUMERICAL mean:-0.00405803 min:-0.110598 max:0.113378 sd:0.0382544 dtype:DTYPE_FLOAT32 2: "embedding.001_of_512" NUMERICAL mean:0.0020755 min:-0.120324 max:0.106003 sd:0.0434171 dtype:DTYPE_FLOAT32 3: "embedding.002_of_512" NUMERICAL mean:0.00763253 min:-0.102393 max:0.126418 sd:0.0391092 dtype:DTYPE_FLOAT32 4: "embedding.003_of_512" NUMERICAL mean:-0.000732218 min:-0.109626 max:0.10121 sd:0.0369311 dtype:DTYPE_FLOAT32 5: "embedding.004_of_512" NUMERICAL mean:0.0126995 min:-0.128462 max:0.120181 sd:0.0460968 dtype:DTYPE_FLOAT32 6: "embedding.005_of_512" NUMERICAL mean:0.0120099 min:-0.120963 max:0.118971 sd:0.041685 dtype:DTYPE_FLOAT32 7: "embedding.006_of_512" NUMERICAL mean:0.00266504 min:-0.107839 max:0.108908 sd:0.0381836 dtype:DTYPE_FLOAT32 8: "embedding.007_of_512" NUMERICAL mean:-0.00274169 min:-0.104539 max:0.13168 sd:0.0381854 dtype:DTYPE_FLOAT32 9: "embedding.008_of_512" NUMERICAL mean:0.0078143 min:-0.1422 max:0.125118 sd:0.0480273 dtype:DTYPE_FLOAT32 10: "embedding.009_of_512" NUMERICAL mean:0.0295029 min:-0.128134 max:0.118291 sd:0.0394542 dtype:DTYPE_FLOAT32 11: "embedding.010_of_512" NUMERICAL mean:0.0143459 min:-0.1118 max:0.118193 sd:0.039633 dtype:DTYPE_FLOAT32 12: "embedding.011_of_512" NUMERICAL mean:7.16191e-05 min:-0.10631 max:0.107239 sd:0.0399338 dtype:DTYPE_FLOAT32 13: "embedding.012_of_512" NUMERICAL mean:-0.0115011 min:-0.115238 max:0.115996 sd:0.039107 dtype:DTYPE_FLOAT32 14: "embedding.013_of_512" NUMERICAL mean:-0.0179365 min:-0.133052 max:0.120982 sd:0.0461472 dtype:DTYPE_FLOAT32 15: "embedding.014_of_512" NUMERICAL mean:0.000907569 min:-0.124092 max:0.111964 sd:0.0393761 dtype:DTYPE_FLOAT32 16: "embedding.015_of_512" NUMERICAL mean:-0.0465314 min:-0.127274 max:0.115007 sd:0.0410307 dtype:DTYPE_FLOAT32 17: "embedding.016_of_512" NUMERICAL mean:0.0122186 min:-0.132531 max:0.112023 sd:0.0412513 dtype:DTYPE_FLOAT32 18: "embedding.017_of_512" NUMERICAL mean:0.0112924 min:-0.118483 max:0.109047 sd:0.0411375 dtype:DTYPE_FLOAT32 19: "embedding.018_of_512" NUMERICAL mean:0.0161367 min:-0.103659 max:0.106467 sd:0.0396646 dtype:DTYPE_FLOAT32 20: "embedding.019_of_512" NUMERICAL mean:0.000475907 min:-0.126049 max:0.109106 sd:0.0416907 dtype:DTYPE_FLOAT32 21: "embedding.020_of_512" NUMERICAL mean:0.0067624 min:-0.117482 max:0.140442 sd:0.0473874 dtype:DTYPE_FLOAT32 22: "embedding.021_of_512" NUMERICAL mean:-0.0103802 min:-0.141084 max:0.11384 sd:0.0543033 dtype:DTYPE_FLOAT32 23: "embedding.022_of_512" NUMERICAL mean:0.00470285 min:-0.108062 max:0.127381 sd:0.0465233 dtype:DTYPE_FLOAT32 24: "embedding.023_of_512" NUMERICAL mean:0.00203851 min:-0.116632 max:0.116226 sd:0.0400387 dtype:DTYPE_FLOAT32 25: "embedding.024_of_512" NUMERICAL mean:-0.000927636 min:-0.127188 max:0.105079 sd:0.042448 dtype:DTYPE_FLOAT32 26: "embedding.025_of_512" NUMERICAL mean:0.0111641 min:-0.114852 max:0.110724 sd:0.0379149 dtype:DTYPE_FLOAT32 27: "embedding.026_of_512" NUMERICAL mean:-0.00327954 min:-0.105336 max:0.104934 sd:0.0414663 dtype:DTYPE_FLOAT32 28: "embedding.027_of_512" NUMERICAL mean:-0.00290058 min:-0.116482 max:0.113443 sd:0.0406192 dtype:DTYPE_FLOAT32 29: "embedding.028_of_512" NUMERICAL mean:0.0028253 min:-0.110413 max:0.123963 sd:0.0427868 dtype:DTYPE_FLOAT32 30: "embedding.029_of_512" NUMERICAL mean:0.0282239 min:-0.104219 max:0.143075 sd:0.0487783 dtype:DTYPE_FLOAT32 31: "embedding.030_of_512" NUMERICAL mean:0.0119399 min:-0.10224 max:0.123741 sd:0.0407582 dtype:DTYPE_FLOAT32 32: "embedding.031_of_512" NUMERICAL mean:-0.00879883 min:-0.106952 max:0.114961 sd:0.0397046 dtype:DTYPE_FLOAT32 33: "embedding.032_of_512" NUMERICAL mean:-0.00641689 min:-0.109096 max:0.115278 sd:0.0395207 dtype:DTYPE_FLOAT32 34: "embedding.033_of_512" NUMERICAL mean:0.00148918 min:-0.111798 max:0.115585 sd:0.0406788 dtype:DTYPE_FLOAT32 35: "embedding.034_of_512" NUMERICAL mean:-0.00225704 min:-0.116123 max:0.116634 sd:0.0410024 dtype:DTYPE_FLOAT32 36: "embedding.035_of_512" NUMERICAL mean:0.0139088 min:-0.11982 max:0.117347 sd:0.0412498 dtype:DTYPE_FLOAT32 37: "embedding.036_of_512" NUMERICAL mean:-0.00414969 min:-0.117693 max:0.10138 sd:0.0421129 dtype:DTYPE_FLOAT32 38: "embedding.037_of_512" NUMERICAL mean:0.00156116 min:-0.114707 max:0.128423 sd:0.0410256 dtype:DTYPE_FLOAT32 39: "embedding.038_of_512" NUMERICAL mean:-0.00506909 min:-0.108533 max:0.113459 sd:0.0408111 dtype:DTYPE_FLOAT32 40: "embedding.039_of_512" NUMERICAL mean:-0.000564469 min:-0.110207 max:0.123467 sd:0.0405413 dtype:DTYPE_FLOAT32 41: "embedding.040_of_512" NUMERICAL mean:0.00390461 min:-0.107059 max:0.128317 sd:0.036992 dtype:DTYPE_FLOAT32 42: "embedding.041_of_512" NUMERICAL mean:-0.00821721 min:-0.124872 max:0.101206 sd:0.0389586 dtype:DTYPE_FLOAT32 43: "embedding.042_of_512" NUMERICAL mean:-0.0263568 min:-0.128555 max:0.12256 sd:0.0438572 dtype:DTYPE_FLOAT32 44: "embedding.043_of_512" NUMERICAL mean:-0.00742056 min:-0.132266 max:0.135953 sd:0.0492916 dtype:DTYPE_FLOAT32 45: "embedding.044_of_512" NUMERICAL mean:-0.0136633 min:-0.125995 max:0.100658 sd:0.0357859 dtype:DTYPE_FLOAT32 46: "embedding.045_of_512" NUMERICAL mean:-0.00672393 min:-0.105793 max:0.106381 sd:0.0395707 dtype:DTYPE_FLOAT32 47: "embedding.046_of_512" NUMERICAL mean:-0.00846792 min:-0.137635 max:0.111598 sd:0.0406422 dtype:DTYPE_FLOAT32 48: "embedding.047_of_512" NUMERICAL mean:0.00542721 min:-0.107879 max:0.112788 sd:0.0407962 dtype:DTYPE_FLOAT32 49: "embedding.048_of_512" NUMERICAL mean:0.00416469 min:-0.106566 max:0.110393 sd:0.0387239 dtype:DTYPE_FLOAT32 50: "embedding.049_of_512" NUMERICAL mean:0.010748 min:-0.101981 max:0.119391 sd:0.0397083 dtype:DTYPE_FLOAT32 51: "embedding.050_of_512" NUMERICAL mean:0.0382225 min:-0.0980938 max:0.129267 sd:0.0373726 dtype:DTYPE_FLOAT32 52: "embedding.051_of_512" NUMERICAL mean:-0.0244919 min:-0.143322 max:0.151466 sd:0.0569238 dtype:DTYPE_FLOAT32 53: "embedding.052_of_512" NUMERICAL mean:0.0137383 min:-0.119567 max:0.149818 sd:0.0480009 dtype:DTYPE_FLOAT32 54: "embedding.053_of_512" NUMERICAL mean:-0.00754001 min:-0.119613 max:0.139327 sd:0.0441231 dtype:DTYPE_FLOAT32 55: "embedding.054_of_512" NUMERICAL mean:-0.00119265 min:-0.117568 max:0.0984011 sd:0.0386896 dtype:DTYPE_FLOAT32 56: "embedding.055_of_512" NUMERICAL mean:-0.00382799 min:-0.113112 max:0.107257 sd:0.0435431 dtype:DTYPE_FLOAT32 57: "embedding.056_of_512" NUMERICAL mean:0.00818074 min:-0.145547 max:0.123275 sd:0.0429192 dtype:DTYPE_FLOAT32 58: "embedding.057_of_512" NUMERICAL mean:-0.00208038 min:-0.126433 max:0.101673 sd:0.0393041 dtype:DTYPE_FLOAT32 59: "embedding.058_of_512" NUMERICAL mean:0.00506083 min:-0.118728 max:0.13801 sd:0.0459501 dtype:DTYPE_FLOAT32 60: "embedding.059_of_512" NUMERICAL mean:-0.00110454 min:-0.111315 max:0.10866 sd:0.0384711 dtype:DTYPE_FLOAT32 61: "embedding.060_of_512" NUMERICAL mean:-0.00560149 min:-0.126673 max:0.142958 sd:0.0476651 dtype:DTYPE_FLOAT32 62: "embedding.061_of_512" NUMERICAL mean:-0.010492 min:-0.116135 max:0.117787 sd:0.0398593 dtype:DTYPE_FLOAT32 63: "embedding.062_of_512" NUMERICAL mean:-0.0196407 min:-0.143423 max:0.104133 sd:0.0483823 dtype:DTYPE_FLOAT32 64: "embedding.063_of_512" NUMERICAL mean:0.0072672 min:-0.134359 max:0.115527 sd:0.0442734 dtype:DTYPE_FLOAT32 65: "embedding.064_of_512" NUMERICAL mean:-0.00813338 min:-0.104328 max:0.11042 sd:0.0378631 dtype:DTYPE_FLOAT32 66: "embedding.065_of_512" NUMERICAL mean:0.0252276 min:-0.134246 max:0.126575 sd:0.0404105 dtype:DTYPE_FLOAT32 67: "embedding.066_of_512" NUMERICAL mean:0.0121496 min:-0.121565 max:0.115153 sd:0.0399014 dtype:DTYPE_FLOAT32 68: "embedding.067_of_512" NUMERICAL mean:0.000328628 min:-0.108976 max:0.10698 sd:0.0409231 dtype:DTYPE_FLOAT32 69: "embedding.068_of_512" NUMERICAL mean:0.0209823 min:-0.111598 max:0.12123 sd:0.0391018 dtype:DTYPE_FLOAT32 70: "embedding.069_of_512" NUMERICAL mean:0.00544792 min:-0.108988 max:0.126124 sd:0.0422695 dtype:DTYPE_FLOAT32 71: "embedding.070_of_512" NUMERICAL mean:-0.000593016 min:-0.119492 max:0.113604 sd:0.0415354 dtype:DTYPE_FLOAT32 72: "embedding.071_of_512" NUMERICAL mean:-0.000604193 min:-0.128741 max:0.107355 sd:0.0426992 dtype:DTYPE_FLOAT32 73: "embedding.072_of_512" NUMERICAL mean:-0.00433507 min:-0.113435 max:0.102836 sd:0.0414469 dtype:DTYPE_FLOAT32 74: "embedding.073_of_512" NUMERICAL mean:-0.0101648 min:-0.10628 max:0.119432 sd:0.0400882 dtype:DTYPE_FLOAT32 75: "embedding.074_of_512" NUMERICAL mean:0.0132994 min:-0.123574 max:0.103854 sd:0.0381882 dtype:DTYPE_FLOAT32 76: "embedding.075_of_512" NUMERICAL mean:-0.00154112 min:-0.135068 max:0.106161 sd:0.0393081 dtype:DTYPE_FLOAT32 77: "embedding.076_of_512" NUMERICAL mean:-0.0107704 min:-0.106198 max:0.106547 sd:0.0380247 dtype:DTYPE_FLOAT32 78: "embedding.077_of_512" NUMERICAL mean:0.0151205 min:-0.0985188 max:0.107297 sd:0.0381537 dtype:DTYPE_FLOAT32 79: "embedding.078_of_512" NUMERICAL mean:0.00829679 min:-0.102936 max:0.116536 sd:0.0410818 dtype:DTYPE_FLOAT32 80: "embedding.079_of_512" NUMERICAL mean:0.00578581 min:-0.156252 max:0.125833 sd:0.0489822 dtype:DTYPE_FLOAT32 81: "embedding.080_of_512" NUMERICAL mean:-0.00466792 min:-0.10975 max:0.118669 sd:0.0422673 dtype:DTYPE_FLOAT32 82: "embedding.081_of_512" NUMERICAL mean:0.00499065 min:-0.0934409 max:0.115151 sd:0.0382445 dtype:DTYPE_FLOAT32 83: "embedding.082_of_512" NUMERICAL mean:-0.0120384 min:-0.115119 max:0.109741 sd:0.039712 dtype:DTYPE_FLOAT32 84: "embedding.083_of_512" NUMERICAL mean:-0.0116498 min:-0.107953 max:0.113206 sd:0.0408114 dtype:DTYPE_FLOAT32 85: "embedding.084_of_512" NUMERICAL mean:-0.0210408 min:-0.108707 max:0.0992159 sd:0.0386516 dtype:DTYPE_FLOAT32 86: "embedding.085_of_512" NUMERICAL mean:-0.00273396 min:-0.12944 max:0.12272 sd:0.0449487 dtype:DTYPE_FLOAT32 87: "embedding.086_of_512" NUMERICAL mean:0.00658216 min:-0.113506 max:0.112219 sd:0.039801 dtype:DTYPE_FLOAT32 88: "embedding.087_of_512" NUMERICAL mean:-0.00378743 min:-0.117676 max:0.109386 sd:0.0402421 dtype:DTYPE_FLOAT32 89: "embedding.088_of_512" NUMERICAL mean:-0.0205237 min:-0.107587 max:0.103141 sd:0.040405 dtype:DTYPE_FLOAT32 90: "embedding.089_of_512" NUMERICAL mean:-0.000411177 min:-0.119937 max:0.109877 sd:0.0421414 dtype:DTYPE_FLOAT32 91: "embedding.090_of_512" NUMERICAL mean:-0.00181531 min:-0.117795 max:0.106343 sd:0.0421115 dtype:DTYPE_FLOAT32 92: "embedding.091_of_512" NUMERICAL mean:-0.00550051 min:-0.127822 max:0.113907 sd:0.0399804 dtype:DTYPE_FLOAT32 93: "embedding.092_of_512" NUMERICAL mean:-0.00547455 min:-0.126723 max:0.119811 sd:0.0431932 dtype:DTYPE_FLOAT32 94: "embedding.093_of_512" NUMERICAL mean:0.014195 min:-0.105489 max:0.118567 sd:0.0413103 dtype:DTYPE_FLOAT32 95: "embedding.094_of_512" NUMERICAL mean:0.0188997 min:-0.104824 max:0.132286 sd:0.0497162 dtype:DTYPE_FLOAT32 96: "embedding.095_of_512" NUMERICAL mean:0.00497901 min:-0.108731 max:0.124192 sd:0.0414468 dtype:DTYPE_FLOAT32 97: "embedding.096_of_512" NUMERICAL mean:-0.0179242 min:-0.125507 max:0.10199 sd:0.0383211 dtype:DTYPE_FLOAT32 98: "embedding.097_of_512" NUMERICAL mean:0.00327183 min:-0.122499 max:0.123037 sd:0.0419092 dtype:DTYPE_FLOAT32 99: "embedding.098_of_512" NUMERICAL mean:0.0216785 min:-0.10081 max:0.116099 sd:0.0479454 dtype:DTYPE_FLOAT32 100: "embedding.099_of_512" NUMERICAL mean:0.019005 min:-0.125922 max:0.117505 sd:0.0429193 dtype:DTYPE_FLOAT32 101: "embedding.100_of_512" NUMERICAL mean:0.003884 min:-0.104019 max:0.127238 sd:0.0431 dtype:DTYPE_FLOAT32 102: "embedding.101_of_512" NUMERICAL mean:-0.0132592 min:-0.133774 max:0.125128 sd:0.0465773 dtype:DTYPE_FLOAT32 103: "embedding.102_of_512" NUMERICAL mean:0.00732224 min:-0.114158 max:0.135181 sd:0.0462208 dtype:DTYPE_FLOAT32 104: "embedding.103_of_512" NUMERICAL mean:-0.00316622 min:-0.115661 max:0.110651 sd:0.0393422 dtype:DTYPE_FLOAT32 105: "embedding.104_of_512" NUMERICAL mean:-0.000406039 min:-0.115186 max:0.115727 sd:0.0404569 dtype:DTYPE_FLOAT32 106: "embedding.105_of_512" NUMERICAL mean:0.01286 min:-0.10478 max:0.116059 sd:0.0408527 dtype:DTYPE_FLOAT32 107: "embedding.106_of_512" NUMERICAL mean:-0.0200857 min:-0.112344 max:0.115696 sd:0.0348447 dtype:DTYPE_FLOAT32 108: "embedding.107_of_512" NUMERICAL mean:-0.0008812 min:-0.117538 max:0.128118 sd:0.0397207 dtype:DTYPE_FLOAT32 109: "embedding.108_of_512" NUMERICAL mean:-0.0153816 min:-0.119853 max:0.111478 sd:0.0408014 dtype:DTYPE_FLOAT32 110: "embedding.109_of_512" NUMERICAL mean:0.0226631 min:-0.115775 max:0.109245 sd:0.0344709 dtype:DTYPE_FLOAT32 111: "embedding.110_of_512" NUMERICAL mean:-0.0117186 min:-0.12628 max:0.0972872 sd:0.043443 dtype:DTYPE_FLOAT32 112: "embedding.111_of_512" NUMERICAL mean:-0.0195 min:-0.138677 max:0.111032 sd:0.0530712 dtype:DTYPE_FLOAT32 113: "embedding.112_of_512" NUMERICAL mean:-0.00883525 min:-0.125434 max:0.115491 sd:0.039556 dtype:DTYPE_FLOAT32 114: "embedding.113_of_512" NUMERICAL mean:-0.0004395 min:-0.106039 max:0.1141 sd:0.0441183 dtype:DTYPE_FLOAT32 115: "embedding.114_of_512" NUMERICAL mean:-0.00404027 min:-0.131798 max:0.106558 sd:0.040391 dtype:DTYPE_FLOAT32 116: "embedding.115_of_512" NUMERICAL mean:0.0164961 min:-0.137229 max:0.11088 sd:0.0396261 dtype:DTYPE_FLOAT32 117: "embedding.116_of_512" NUMERICAL mean:-0.0163338 min:-0.109692 max:0.115104 sd:0.0396108 dtype:DTYPE_FLOAT32 118: "embedding.117_of_512" NUMERICAL mean:-0.000866382 min:-0.111258 max:0.110021 sd:0.0413076 dtype:DTYPE_FLOAT32 119: "embedding.118_of_512" NUMERICAL mean:0.00925641 min:-0.117275 max:0.109073 sd:0.0392531 dtype:DTYPE_FLOAT32 120: "embedding.119_of_512" NUMERICAL mean:0.0111224 min:-0.108271 max:0.11018 sd:0.0438516 dtype:DTYPE_FLOAT32 121: "embedding.120_of_512" NUMERICAL mean:-0.0109583 min:-0.117243 max:0.113314 sd:0.03753 dtype:DTYPE_FLOAT32 122: "embedding.121_of_512" NUMERICAL mean:0.0143342 min:-0.109885 max:0.121471 sd:0.0401907 dtype:DTYPE_FLOAT32 123: "embedding.122_of_512" NUMERICAL mean:-0.00603129 min:-0.111126 max:0.106422 sd:0.0401383 dtype:DTYPE_FLOAT32 124: "embedding.123_of_512" NUMERICAL mean:-0.00175511 min:-0.115219 max:0.103571 sd:0.0388962 dtype:DTYPE_FLOAT32 125: "embedding.124_of_512" NUMERICAL mean:-0.0119755 min:-0.119062 max:0.122632 sd:0.0447561 dtype:DTYPE_FLOAT32 126: "embedding.125_of_512" NUMERICAL mean:0.00210507 min:-0.116783 max:0.125758 sd:0.0469827 dtype:DTYPE_FLOAT32 127: "embedding.126_of_512" NUMERICAL mean:-0.0166424 min:-0.109771 max:0.13027 sd:0.0399639 dtype:DTYPE_FLOAT32 128: "embedding.127_of_512" NUMERICAL mean:-0.0462275 min:-0.137916 max:0.106133 sd:0.0478679 dtype:DTYPE_FLOAT32 129: "embedding.128_of_512" NUMERICAL mean:0.0101449 min:-0.134851 max:0.118003 sd:0.0415072 dtype:DTYPE_FLOAT32 130: "embedding.129_of_512" NUMERICAL mean:0.0119622 min:-0.106398 max:0.122529 sd:0.047894 dtype:DTYPE_FLOAT32 131: "embedding.130_of_512" NUMERICAL mean:-0.0109302 min:-0.127096 max:0.102555 sd:0.0407236 dtype:DTYPE_FLOAT32 132: "embedding.131_of_512" NUMERICAL mean:-2.30423e-05 min:-0.0958128 max:0.116109 sd:0.0393919 dtype:DTYPE_FLOAT32 133: "embedding.132_of_512" NUMERICAL mean:0.00622466 min:-0.118524 max:0.171935 sd:0.0435631 dtype:DTYPE_FLOAT32 134: "embedding.133_of_512" NUMERICAL mean:0.00537511 min:-0.0999398 max:0.143991 sd:0.0431652 dtype:DTYPE_FLOAT32 135: "embedding.134_of_512" NUMERICAL mean:0.0111946 min:-0.101547 max:0.105716 sd:0.0365295 dtype:DTYPE_FLOAT32 136: "embedding.135_of_512" NUMERICAL mean:-0.0123165 min:-0.118347 max:0.113619 sd:0.0422525 dtype:DTYPE_FLOAT32 137: "embedding.136_of_512" NUMERICAL mean:0.00882626 min:-0.118642 max:0.115052 sd:0.0393646 dtype:DTYPE_FLOAT32 138: "embedding.137_of_512" NUMERICAL mean:0.0106701 min:-0.108036 max:0.109746 sd:0.0405698 dtype:DTYPE_FLOAT32 139: "embedding.138_of_512" NUMERICAL mean:-0.0130655 min:-0.148064 max:0.118745 sd:0.047092 dtype:DTYPE_FLOAT32 140: "embedding.139_of_512" NUMERICAL mean:0.00256777 min:-0.108547 max:0.102547 sd:0.0388182 dtype:DTYPE_FLOAT32 141: "embedding.140_of_512" NUMERICAL mean:-0.00255201 min:-0.113298 max:0.120327 sd:0.0469564 dtype:DTYPE_FLOAT32 142: "embedding.141_of_512" NUMERICAL mean:-0.0123127 min:-0.124039 max:0.110528 sd:0.047218 dtype:DTYPE_FLOAT32 143: "embedding.142_of_512" NUMERICAL mean:0.00659571 min:-0.106909 max:0.126327 sd:0.0444828 dtype:DTYPE_FLOAT32 144: "embedding.143_of_512" NUMERICAL mean:0.00838607 min:-0.121819 max:0.108286 sd:0.0409403 dtype:DTYPE_FLOAT32 145: "embedding.144_of_512" NUMERICAL mean:-0.00504916 min:-0.117741 max:0.109832 sd:0.0402179 dtype:DTYPE_FLOAT32 146: "embedding.145_of_512" NUMERICAL mean:-0.0135 min:-0.112358 max:0.108238 sd:0.0393695 dtype:DTYPE_FLOAT32 147: "embedding.146_of_512" NUMERICAL mean:-0.00551706 min:-0.108132 max:0.103118 sd:0.0375181 dtype:DTYPE_FLOAT32 148: "embedding.147_of_512" NUMERICAL mean:0.00226707 min:-0.109358 max:0.117688 sd:0.0416268 dtype:DTYPE_FLOAT32 149: "embedding.148_of_512" NUMERICAL mean:-0.0083477 min:-0.113886 max:0.105174 sd:0.0379074 dtype:DTYPE_FLOAT32 150: "embedding.149_of_512" NUMERICAL mean:-0.0029158 min:-0.104327 max:0.10898 sd:0.0394245 dtype:DTYPE_FLOAT32 151: "embedding.150_of_512" NUMERICAL mean:-0.00857055 min:-0.11757 max:0.108206 sd:0.0416898 dtype:DTYPE_FLOAT32 152: "embedding.151_of_512" NUMERICAL mean:0.00697777 min:-0.104269 max:0.109967 sd:0.0353302 dtype:DTYPE_FLOAT32 153: "embedding.152_of_512" NUMERICAL mean:-0.0220037 min:-0.122602 max:0.105503 sd:0.0429071 dtype:DTYPE_FLOAT32 154: "embedding.153_of_512" NUMERICAL mean:-0.00103943 min:-0.109326 max:0.112115 sd:0.0413219 dtype:DTYPE_FLOAT32 155: "embedding.154_of_512" NUMERICAL mean:-0.010306 min:-0.106116 max:0.112623 sd:0.0392094 dtype:DTYPE_FLOAT32 156: "embedding.155_of_512" NUMERICAL mean:-0.0128503 min:-0.133511 max:0.129721 sd:0.0417087 dtype:DTYPE_FLOAT32 157: "embedding.156_of_512" NUMERICAL mean:-0.00796017 min:-0.10801 max:0.111555 sd:0.0401771 dtype:DTYPE_FLOAT32 158: "embedding.157_of_512" NUMERICAL mean:-0.0263644 min:-0.135057 max:0.131898 sd:0.0473006 dtype:DTYPE_FLOAT32 159: "embedding.158_of_512" NUMERICAL mean:0.0157188 min:-0.109795 max:0.13194 sd:0.0423631 dtype:DTYPE_FLOAT32 160: "embedding.159_of_512" NUMERICAL mean:0.00616692 min:-0.0996693 max:0.121898 sd:0.0405747 dtype:DTYPE_FLOAT32 161: "embedding.160_of_512" NUMERICAL mean:0.00140896 min:-0.125797 max:0.10415 sd:0.0422833 dtype:DTYPE_FLOAT32 162: "embedding.161_of_512" NUMERICAL mean:-0.00968098 min:-0.107129 max:0.109673 sd:0.0389125 dtype:DTYPE_FLOAT32 163: "embedding.162_of_512" NUMERICAL mean:0.0174977 min:-0.102559 max:0.117249 sd:0.0394065 dtype:DTYPE_FLOAT32 164: "embedding.163_of_512" NUMERICAL mean:-0.01559 min:-0.117529 max:0.132716 sd:0.0422287 dtype:DTYPE_FLOAT32 165: "embedding.164_of_512" NUMERICAL mean:0.0103332 min:-0.131635 max:0.117116 sd:0.0432647 dtype:DTYPE_FLOAT32 166: "embedding.165_of_512" NUMERICAL mean:0.0164754 min:-0.111395 max:0.106868 sd:0.03591 dtype:DTYPE_FLOAT32 167: "embedding.166_of_512" NUMERICAL mean:-0.0300909 min:-0.110079 max:0.138071 sd:0.0393771 dtype:DTYPE_FLOAT32 168: "embedding.167_of_512" NUMERICAL mean:-0.00284721 min:-0.113047 max:0.1113 sd:0.0402787 dtype:DTYPE_FLOAT32 169: "embedding.168_of_512" NUMERICAL mean:0.0128449 min:-0.123295 max:0.101678 sd:0.035443 dtype:DTYPE_FLOAT32 170: "embedding.169_of_512" NUMERICAL mean:-0.0018307 min:-0.113497 max:0.108755 sd:0.0385736 dtype:DTYPE_FLOAT32 171: "embedding.170_of_512" NUMERICAL mean:-0.0154471 min:-0.123997 max:0.0995884 sd:0.039095 dtype:DTYPE_FLOAT32 172: "embedding.171_of_512" NUMERICAL mean:-0.0115266 min:-0.135629 max:0.111586 sd:0.0564499 dtype:DTYPE_FLOAT32 173: "embedding.172_of_512" NUMERICAL mean:-0.00305818 min:-0.108148 max:0.125287 sd:0.0416153 dtype:DTYPE_FLOAT32 174: "embedding.173_of_512" NUMERICAL mean:-0.0192183 min:-0.128661 max:0.111586 sd:0.0445312 dtype:DTYPE_FLOAT32 175: "embedding.174_of_512" NUMERICAL mean:-0.00547071 min:-0.106778 max:0.107318 sd:0.0412694 dtype:DTYPE_FLOAT32 176: "embedding.175_of_512" NUMERICAL mean:0.00303105 min:-0.114183 max:0.11671 sd:0.037753 dtype:DTYPE_FLOAT32 177: "embedding.176_of_512" NUMERICAL mean:0.0200632 min:-0.119154 max:0.12262 sd:0.0449386 dtype:DTYPE_FLOAT32 178: "embedding.177_of_512" NUMERICAL mean:0.00830421 min:-0.106867 max:0.108159 sd:0.04212 dtype:DTYPE_FLOAT32 179: "embedding.178_of_512" NUMERICAL mean:0.00879771 min:-0.119236 max:0.0975505 sd:0.0365596 dtype:DTYPE_FLOAT32 180: "embedding.179_of_512" NUMERICAL mean:-0.0224472 min:-0.141699 max:0.121597 sd:0.0451563 dtype:DTYPE_FLOAT32 181: "embedding.180_of_512" NUMERICAL mean:0.00700458 min:-0.122243 max:0.106828 sd:0.0406674 dtype:DTYPE_FLOAT32 182: "embedding.181_of_512" NUMERICAL mean:0.015665 min:-0.123784 max:0.117493 sd:0.0423638 dtype:DTYPE_FLOAT32 183: "embedding.182_of_512" NUMERICAL mean:0.00455087 min:-0.130433 max:0.129947 sd:0.0468312 dtype:DTYPE_FLOAT32 184: "embedding.183_of_512" NUMERICAL mean:0.00469912 min:-0.105513 max:0.115268 sd:0.0422015 dtype:DTYPE_FLOAT32 185: "embedding.184_of_512" NUMERICAL mean:0.00118913 min:-0.132085 max:0.119005 sd:0.0425006 dtype:DTYPE_FLOAT32 186: "embedding.185_of_512" NUMERICAL mean:-0.0091211 min:-0.105384 max:0.107321 sd:0.0394833 dtype:DTYPE_FLOAT32 187: "embedding.186_of_512" NUMERICAL mean:0.00847289 min:-0.100142 max:0.11416 sd:0.0354507 dtype:DTYPE_FLOAT32 188: "embedding.187_of_512" NUMERICAL mean:0.00401229 min:-0.0997345 max:0.0985512 sd:0.0330015 dtype:DTYPE_FLOAT32 189: "embedding.188_of_512" NUMERICAL mean:0.0375059 min:-0.107009 max:0.147423 sd:0.0457626 dtype:DTYPE_FLOAT32 190: "embedding.189_of_512" NUMERICAL mean:-0.0108558 min:-0.158798 max:0.124698 sd:0.0429543 dtype:DTYPE_FLOAT32 191: "embedding.190_of_512" NUMERICAL mean:0.0055649 min:-0.102637 max:0.112907 sd:0.0428818 dtype:DTYPE_FLOAT32 192: "embedding.191_of_512" NUMERICAL mean:0.0115727 min:-0.0992453 max:0.114756 sd:0.0385606 dtype:DTYPE_FLOAT32 193: "embedding.192_of_512" NUMERICAL mean:0.0188207 min:-0.10799 max:0.126446 sd:0.0480458 dtype:DTYPE_FLOAT32 194: "embedding.193_of_512" NUMERICAL mean:-0.0231128 min:-0.125829 max:0.098485 sd:0.0413616 dtype:DTYPE_FLOAT32 195: "embedding.194_of_512" NUMERICAL mean:-0.0125518 min:-0.118983 max:0.111524 sd:0.0394032 dtype:DTYPE_FLOAT32 196: "embedding.195_of_512" NUMERICAL mean:-0.00734374 min:-0.140773 max:0.124731 sd:0.048662 dtype:DTYPE_FLOAT32 197: "embedding.196_of_512" NUMERICAL mean:0.0147101 min:-0.109208 max:0.114207 sd:0.0392372 dtype:DTYPE_FLOAT32 198: "embedding.197_of_512" NUMERICAL mean:0.00382817 min:-0.0960263 max:0.109744 sd:0.0343786 dtype:DTYPE_FLOAT32 199: "embedding.198_of_512" NUMERICAL mean:0.0148358 min:-0.121261 max:0.137886 sd:0.0396124 dtype:DTYPE_FLOAT32 200: "embedding.199_of_512" NUMERICAL mean:0.0139377 min:-0.133057 max:0.129123 sd:0.0434494 dtype:DTYPE_FLOAT32 201: "embedding.200_of_512" NUMERICAL mean:-0.022174 min:-0.135182 max:0.0998059 sd:0.0447171 dtype:DTYPE_FLOAT32 202: "embedding.201_of_512" NUMERICAL mean:0.00918432 min:-0.129768 max:0.104146 sd:0.0407455 dtype:DTYPE_FLOAT32 203: "embedding.202_of_512" NUMERICAL mean:6.68976e-05 min:-0.108528 max:0.112123 sd:0.039669 dtype:DTYPE_FLOAT32 204: "embedding.203_of_512" NUMERICAL mean:-0.0211792 min:-0.138447 max:0.151201 sd:0.0475548 dtype:DTYPE_FLOAT32 205: "embedding.204_of_512" NUMERICAL mean:0.0149458 min:-0.114192 max:0.121993 sd:0.0451805 dtype:DTYPE_FLOAT32 206: "embedding.205_of_512" NUMERICAL mean:-0.000877425 min:-0.106281 max:0.110069 sd:0.0399283 dtype:DTYPE_FLOAT32 207: "embedding.206_of_512" NUMERICAL mean:0.00135042 min:-0.122458 max:0.133155 sd:0.0490798 dtype:DTYPE_FLOAT32 208: "embedding.207_of_512" NUMERICAL mean:-0.00564687 min:-0.0980346 max:0.124534 sd:0.0381495 dtype:DTYPE_FLOAT32 209: "embedding.208_of_512" NUMERICAL mean:-0.0137386 min:-0.104712 max:0.116268 sd:0.0380542 dtype:DTYPE_FLOAT32 210: "embedding.209_of_512" NUMERICAL mean:-0.000932724 min:-0.120575 max:0.106782 sd:0.0389735 dtype:DTYPE_FLOAT32 211: "embedding.210_of_512" NUMERICAL mean:-0.0221436 min:-0.11615 max:0.110612 sd:0.0375885 dtype:DTYPE_FLOAT32 212: "embedding.211_of_512" NUMERICAL mean:0.00739621 min:-0.107881 max:0.139283 sd:0.0380559 dtype:DTYPE_FLOAT32 213: "embedding.212_of_512" NUMERICAL mean:0.000771754 min:-0.130277 max:0.118151 sd:0.0457612 dtype:DTYPE_FLOAT32 214: "embedding.213_of_512" NUMERICAL mean:-0.00631693 min:-0.113811 max:0.122369 sd:0.0420019 dtype:DTYPE_FLOAT32 215: "embedding.214_of_512" NUMERICAL mean:-0.0190752 min:-0.130814 max:0.12256 sd:0.0462656 dtype:DTYPE_FLOAT32 216: "embedding.215_of_512" NUMERICAL mean:0.00351438 min:-0.119497 max:0.112531 sd:0.0389063 dtype:DTYPE_FLOAT32 217: "embedding.216_of_512" NUMERICAL mean:-0.00563816 min:-0.113327 max:0.108573 sd:0.0398438 dtype:DTYPE_FLOAT32 218: "embedding.217_of_512" NUMERICAL mean:-0.0128165 min:-0.152494 max:0.112129 sd:0.0435284 dtype:DTYPE_FLOAT32 219: "embedding.218_of_512" NUMERICAL mean:-0.000746105 min:-0.115932 max:0.103357 sd:0.0396475 dtype:DTYPE_FLOAT32 220: "embedding.219_of_512" NUMERICAL mean:0.00706257 min:-0.105737 max:0.115808 sd:0.0415758 dtype:DTYPE_FLOAT32 221: "embedding.220_of_512" NUMERICAL mean:0.000614337 min:-0.120866 max:0.10502 sd:0.036915 dtype:DTYPE_FLOAT32 222: "embedding.221_of_512" NUMERICAL mean:-0.00315481 min:-0.110209 max:0.126778 sd:0.0398762 dtype:DTYPE_FLOAT32 223: "embedding.222_of_512" NUMERICAL mean:-0.0055338 min:-0.112973 max:0.111057 sd:0.0367833 dtype:DTYPE_FLOAT32 224: "embedding.223_of_512" NUMERICAL mean:0.0129532 min:-0.108908 max:0.112232 sd:0.0406737 dtype:DTYPE_FLOAT32 225: "embedding.224_of_512" NUMERICAL mean:-0.0195448 min:-0.112833 max:0.122565 sd:0.0423641 dtype:DTYPE_FLOAT32 226: "embedding.225_of_512" NUMERICAL mean:0.00715641 min:-0.136763 max:0.123146 sd:0.0455536 dtype:DTYPE_FLOAT32 227: "embedding.226_of_512" NUMERICAL mean:0.0105978 min:-0.121166 max:0.125465 sd:0.0433322 dtype:DTYPE_FLOAT32 228: "embedding.227_of_512" NUMERICAL mean:-0.00822156 min:-0.131487 max:0.125193 sd:0.0440489 dtype:DTYPE_FLOAT32 229: "embedding.228_of_512" NUMERICAL mean:0.0119113 min:-0.109956 max:0.107868 sd:0.0382855 dtype:DTYPE_FLOAT32 230: "embedding.229_of_512" NUMERICAL mean:-0.00739044 min:-0.116468 max:0.109886 sd:0.0406385 dtype:DTYPE_FLOAT32 231: "embedding.230_of_512" NUMERICAL mean:0.00819752 min:-0.100016 max:0.125019 sd:0.041894 dtype:DTYPE_FLOAT32 232: "embedding.231_of_512" NUMERICAL mean:-0.00420582 min:-0.139816 max:0.138647 sd:0.0446602 dtype:DTYPE_FLOAT32 233: "embedding.232_of_512" NUMERICAL mean:0.00810722 min:-0.11301 max:0.106853 sd:0.0400325 dtype:DTYPE_FLOAT32 234: "embedding.233_of_512" NUMERICAL mean:0.0561205 min:-0.110581 max:0.182054 sd:0.0645425 dtype:DTYPE_FLOAT32 235: "embedding.234_of_512" NUMERICAL mean:0.0202212 min:-0.109987 max:0.116563 sd:0.0374199 dtype:DTYPE_FLOAT32 236: "embedding.235_of_512" NUMERICAL mean:-0.0125547 min:-0.104766 max:0.115993 sd:0.0383767 dtype:DTYPE_FLOAT32 237: "embedding.236_of_512" NUMERICAL mean:0.00228544 min:-0.126092 max:0.125991 sd:0.0403744 dtype:DTYPE_FLOAT32 238: "embedding.237_of_512" NUMERICAL mean:-0.00306858 min:-0.107907 max:0.109284 sd:0.0409564 dtype:DTYPE_FLOAT32 239: "embedding.238_of_512" NUMERICAL mean:-0.00930815 min:-0.156445 max:0.107558 sd:0.0437983 dtype:DTYPE_FLOAT32 240: "embedding.239_of_512" NUMERICAL mean:0.00958206 min:-0.112118 max:0.1195 sd:0.0451739 dtype:DTYPE_FLOAT32 241: "embedding.240_of_512" NUMERICAL mean:-0.00998686 min:-0.125181 max:0.107936 sd:0.0414998 dtype:DTYPE_FLOAT32 242: "embedding.241_of_512" NUMERICAL mean:-0.00128156 min:-0.103688 max:0.109599 sd:0.0377828 dtype:DTYPE_FLOAT32 243: "embedding.242_of_512" NUMERICAL mean:-0.000524396 min:-0.141003 max:0.114016 sd:0.050088 dtype:DTYPE_FLOAT32 244: "embedding.243_of_512" NUMERICAL mean:-0.000359091 min:-0.114483 max:0.130721 sd:0.0418654 dtype:DTYPE_FLOAT32 245: "embedding.244_of_512" NUMERICAL mean:0.0161613 min:-0.103932 max:0.116754 sd:0.0401808 dtype:DTYPE_FLOAT32 246: "embedding.245_of_512" NUMERICAL mean:0.0275608 min:-0.127227 max:0.143614 sd:0.0465002 dtype:DTYPE_FLOAT32 247: "embedding.246_of_512" NUMERICAL mean:-0.0199729 min:-0.107911 max:0.114303 sd:0.037755 dtype:DTYPE_FLOAT32 248: "embedding.247_of_512" NUMERICAL mean:-0.00782877 min:-0.104362 max:0.11543 sd:0.041834 dtype:DTYPE_FLOAT32 249: "embedding.248_of_512" NUMERICAL mean:-0.00054477 min:-0.159329 max:0.155847 sd:0.0543164 dtype:DTYPE_FLOAT32 250: "embedding.249_of_512" NUMERICAL mean:-0.0101255 min:-0.116432 max:0.107342 sd:0.0401119 dtype:DTYPE_FLOAT32 251: "embedding.250_of_512" NUMERICAL mean:0.0161291 min:-0.12229 max:0.109533 sd:0.0372791 dtype:DTYPE_FLOAT32 252: "embedding.251_of_512" NUMERICAL mean:-0.000411384 min:-0.118338 max:0.116215 sd:0.0459737 dtype:DTYPE_FLOAT32 253: "embedding.252_of_512" NUMERICAL mean:-0.00268351 min:-0.108327 max:0.109842 sd:0.037631 dtype:DTYPE_FLOAT32 254: "embedding.253_of_512" NUMERICAL mean:-0.00246653 min:-0.107393 max:0.114115 sd:0.0386872 dtype:DTYPE_FLOAT32 255: "embedding.254_of_512" NUMERICAL mean:0.00223856 min:-0.122731 max:0.140702 sd:0.0447316 dtype:DTYPE_FLOAT32 256: "embedding.255_of_512" NUMERICAL mean:0.00186748 min:-0.128662 max:0.107003 sd:0.0409741 dtype:DTYPE_FLOAT32 257: "embedding.256_of_512" NUMERICAL mean:0.00786944 min:-0.113685 max:0.118287 sd:0.0418721 dtype:DTYPE_FLOAT32 258: "embedding.257_of_512" NUMERICAL mean:-0.00450053 min:-0.117383 max:0.138567 sd:0.0535368 dtype:DTYPE_FLOAT32 259: "embedding.258_of_512" NUMERICAL mean:0.0128997 min:-0.109905 max:0.118147 sd:0.0393103 dtype:DTYPE_FLOAT32 260: "embedding.259_of_512" NUMERICAL mean:0.00794854 min:-0.10424 max:0.111261 sd:0.0380286 dtype:DTYPE_FLOAT32 261: "embedding.260_of_512" NUMERICAL mean:-0.0117858 min:-0.116906 max:0.103426 sd:0.0360148 dtype:DTYPE_FLOAT32 262: "embedding.261_of_512" NUMERICAL mean:0.00338883 min:-0.113532 max:0.114904 sd:0.0432436 dtype:DTYPE_FLOAT32 263: "embedding.262_of_512" NUMERICAL mean:0.00238206 min:-0.149582 max:0.13639 sd:0.0507155 dtype:DTYPE_FLOAT32 264: "embedding.263_of_512" NUMERICAL mean:-0.0103074 min:-0.140884 max:0.117382 sd:0.0508164 dtype:DTYPE_FLOAT32 265: "embedding.264_of_512" NUMERICAL mean:0.00478302 min:-0.104717 max:0.125411 sd:0.0411592 dtype:DTYPE_FLOAT32 266: "embedding.265_of_512" NUMERICAL mean:0.00418632 min:-0.111659 max:0.125069 sd:0.0400184 dtype:DTYPE_FLOAT32 267: "embedding.266_of_512" NUMERICAL mean:-0.0065648 min:-0.115424 max:0.115422 sd:0.040284 dtype:DTYPE_FLOAT32 268: "embedding.267_of_512" NUMERICAL mean:-0.0108974 min:-0.140032 max:0.108537 sd:0.0416651 dtype:DTYPE_FLOAT32 269: "embedding.268_of_512" NUMERICAL mean:0.021397 min:-0.110922 max:0.120673 sd:0.0416704 dtype:DTYPE_FLOAT32 270: "embedding.269_of_512" NUMERICAL mean:-0.00266875 min:-0.108534 max:0.116014 sd:0.0454318 dtype:DTYPE_FLOAT32 271: "embedding.270_of_512" NUMERICAL mean:0.00904486 min:-0.130418 max:0.158166 sd:0.0548252 dtype:DTYPE_FLOAT32 272: "embedding.271_of_512" NUMERICAL mean:0.00193987 min:-0.137558 max:0.14649 sd:0.0508115 dtype:DTYPE_FLOAT32 273: "embedding.272_of_512" NUMERICAL mean:-0.000186977 min:-0.116413 max:0.0989802 sd:0.0402487 dtype:DTYPE_FLOAT32 274: "embedding.273_of_512" NUMERICAL mean:0.006326 min:-0.115043 max:0.107482 sd:0.0416155 dtype:DTYPE_FLOAT32 275: "embedding.274_of_512" NUMERICAL mean:-0.000278916 min:-0.115695 max:0.105325 sd:0.0406986 dtype:DTYPE_FLOAT32 276: "embedding.275_of_512" NUMERICAL mean:-0.0102959 min:-0.099434 max:0.128947 sd:0.0361354 dtype:DTYPE_FLOAT32 277: "embedding.276_of_512" NUMERICAL mean:-0.0207918 min:-0.116139 max:0.110566 sd:0.0419115 dtype:DTYPE_FLOAT32 278: "embedding.277_of_512" NUMERICAL mean:-0.0146824 min:-0.127741 max:0.101543 sd:0.0430422 dtype:DTYPE_FLOAT32 279: "embedding.278_of_512" NUMERICAL mean:0.0187157 min:-0.109012 max:0.119525 sd:0.0469243 dtype:DTYPE_FLOAT32 280: "embedding.279_of_512" NUMERICAL mean:0.0080616 min:-0.117272 max:0.138517 sd:0.0500966 dtype:DTYPE_FLOAT32 281: "embedding.280_of_512" NUMERICAL mean:0.0017946 min:-0.129883 max:0.103422 sd:0.0466893 dtype:DTYPE_FLOAT32 282: "embedding.281_of_512" NUMERICAL mean:-0.00546588 min:-0.123351 max:0.122337 sd:0.044948 dtype:DTYPE_FLOAT32 283: "embedding.282_of_512" NUMERICAL mean:-0.00352354 min:-0.114364 max:0.122504 sd:0.0421913 dtype:DTYPE_FLOAT32 284: "embedding.283_of_512" NUMERICAL mean:0.00593286 min:-0.104898 max:0.11458 sd:0.0418491 dtype:DTYPE_FLOAT32 285: "embedding.284_of_512" NUMERICAL mean:-0.0136068 min:-0.112147 max:0.110563 sd:0.0402539 dtype:DTYPE_FLOAT32 286: "embedding.285_of_512" NUMERICAL mean:-0.0148682 min:-0.143126 max:0.121947 sd:0.0652969 dtype:DTYPE_FLOAT32 287: "embedding.286_of_512" NUMERICAL mean:0.00865603 min:-0.105883 max:0.116117 sd:0.0411941 dtype:DTYPE_FLOAT32 288: "embedding.287_of_512" NUMERICAL mean:0.00838776 min:-0.103808 max:0.118732 sd:0.0400033 dtype:DTYPE_FLOAT32 289: "embedding.288_of_512" NUMERICAL mean:-0.004587 min:-0.126515 max:0.110044 sd:0.0429655 dtype:DTYPE_FLOAT32 290: "embedding.289_of_512" NUMERICAL mean:0.022459 min:-0.101127 max:0.122341 sd:0.0412413 dtype:DTYPE_FLOAT32 291: "embedding.290_of_512" NUMERICAL mean:-0.010227 min:-0.104646 max:0.11767 sd:0.0391759 dtype:DTYPE_FLOAT32 292: "embedding.291_of_512" NUMERICAL mean:0.0479376 min:-0.118972 max:0.140115 sd:0.0441115 dtype:DTYPE_FLOAT32 293: "embedding.292_of_512" NUMERICAL mean:-0.012885 min:-0.13523 max:0.1102 sd:0.044191 dtype:DTYPE_FLOAT32 294: "embedding.293_of_512" NUMERICAL mean:-0.00582894 min:-0.118518 max:0.1084 sd:0.0424979 dtype:DTYPE_FLOAT32 295: "embedding.294_of_512" NUMERICAL mean:0.00673141 min:-0.123867 max:0.135324 sd:0.0469895 dtype:DTYPE_FLOAT32 296: "embedding.295_of_512" NUMERICAL mean:0.00592276 min:-0.109027 max:0.121098 sd:0.0376266 dtype:DTYPE_FLOAT32 297: "embedding.296_of_512" NUMERICAL mean:-0.000323969 min:-0.132564 max:0.106466 sd:0.0429391 dtype:DTYPE_FLOAT32 298: "embedding.297_of_512" NUMERICAL mean:0.00159954 min:-0.10937 max:0.112449 sd:0.0405972 dtype:DTYPE_FLOAT32 299: "embedding.298_of_512" NUMERICAL mean:0.0203997 min:-0.130037 max:0.102531 sd:0.0376077 dtype:DTYPE_FLOAT32 300: "embedding.299_of_512" NUMERICAL mean:0.00443814 min:-0.126552 max:0.0985593 sd:0.0406299 dtype:DTYPE_FLOAT32 301: "embedding.300_of_512" NUMERICAL mean:0.00640362 min:-0.109722 max:0.113832 sd:0.042602 dtype:DTYPE_FLOAT32 302: "embedding.301_of_512" NUMERICAL mean:0.00300331 min:-0.10537 max:0.1057 sd:0.0400365 dtype:DTYPE_FLOAT32 303: "embedding.302_of_512" NUMERICAL mean:0.0105726 min:-0.125406 max:0.125337 sd:0.0386879 dtype:DTYPE_FLOAT32 304: "embedding.303_of_512" NUMERICAL mean:-0.00682487 min:-0.119722 max:0.122495 sd:0.0397744 dtype:DTYPE_FLOAT32 305: "embedding.304_of_512" NUMERICAL mean:0.0134615 min:-0.113637 max:0.104308 sd:0.0364568 dtype:DTYPE_FLOAT32 306: "embedding.305_of_512" NUMERICAL mean:0.00644908 min:-0.106984 max:0.118193 sd:0.0378877 dtype:DTYPE_FLOAT32 307: "embedding.306_of_512" NUMERICAL mean:0.00721292 min:-0.106136 max:0.112877 sd:0.0413748 dtype:DTYPE_FLOAT32 308: "embedding.307_of_512" NUMERICAL mean:-0.00382715 min:-0.104953 max:0.0990278 sd:0.0384972 dtype:DTYPE_FLOAT32 309: "embedding.308_of_512" NUMERICAL mean:6.43181e-05 min:-0.120151 max:0.118558 sd:0.0443767 dtype:DTYPE_FLOAT32 310: "embedding.309_of_512" NUMERICAL mean:0.00712577 min:-0.118636 max:0.108645 sd:0.0429865 dtype:DTYPE_FLOAT32 311: "embedding.310_of_512" NUMERICAL mean:0.00668597 min:-0.10649 max:0.116392 sd:0.040195 dtype:DTYPE_FLOAT32 312: "embedding.311_of_512" NUMERICAL mean:-0.000903381 min:-0.119513 max:0.131158 sd:0.0420348 dtype:DTYPE_FLOAT32 313: "embedding.312_of_512" NUMERICAL mean:0.0107332 min:-0.113776 max:0.112523 sd:0.0408102 dtype:DTYPE_FLOAT32 314: "embedding.313_of_512" NUMERICAL mean:0.00918225 min:-0.103286 max:0.106814 sd:0.03942 dtype:DTYPE_FLOAT32 315: "embedding.314_of_512" NUMERICAL mean:0.00465102 min:-0.110279 max:0.117252 sd:0.0393894 dtype:DTYPE_FLOAT32 316: "embedding.315_of_512" NUMERICAL mean:-0.00789822 min:-0.107114 max:0.11401 sd:0.0388347 dtype:DTYPE_FLOAT32 317: "embedding.316_of_512" NUMERICAL mean:0.003646 min:-0.115399 max:0.102757 sd:0.0402218 dtype:DTYPE_FLOAT32 318: "embedding.317_of_512" NUMERICAL mean:0.015828 min:-0.115321 max:0.130694 sd:0.0440749 dtype:DTYPE_FLOAT32 319: "embedding.318_of_512" NUMERICAL mean:-0.0205412 min:-0.115586 max:0.144723 sd:0.0485943 dtype:DTYPE_FLOAT32 320: "embedding.319_of_512" NUMERICAL mean:0.00661137 min:-0.121465 max:0.11194 sd:0.0411842 dtype:DTYPE_FLOAT32 321: "embedding.320_of_512" NUMERICAL mean:-0.0148287 min:-0.103164 max:0.116781 sd:0.0390764 dtype:DTYPE_FLOAT32 322: "embedding.321_of_512" NUMERICAL mean:-0.0216578 min:-0.124605 max:0.115269 sd:0.0434055 dtype:DTYPE_FLOAT32 323: "embedding.322_of_512" NUMERICAL mean:0.00985385 min:-0.100306 max:0.1268 sd:0.0390696 dtype:DTYPE_FLOAT32 324: "embedding.323_of_512" NUMERICAL mean:0.00628717 min:-0.0997496 max:0.119355 sd:0.0396103 dtype:DTYPE_FLOAT32 325: "embedding.324_of_512" NUMERICAL mean:-0.00196284 min:-0.121922 max:0.120337 sd:0.0459949 dtype:DTYPE_FLOAT32 326: "embedding.325_of_512" NUMERICAL mean:-0.00537022 min:-0.110575 max:0.123165 sd:0.0455996 dtype:DTYPE_FLOAT32 327: "embedding.326_of_512" NUMERICAL mean:0.00455174 min:-0.115791 max:0.104665 sd:0.0401681 dtype:DTYPE_FLOAT32 328: "embedding.327_of_512" NUMERICAL mean:-0.00533296 min:-0.130506 max:0.112283 sd:0.0453555 dtype:DTYPE_FLOAT32 329: "embedding.328_of_512" NUMERICAL mean:-0.00440578 min:-0.126272 max:0.103891 sd:0.041464 dtype:DTYPE_FLOAT32 330: "embedding.329_of_512" NUMERICAL mean:-0.0101936 min:-0.108874 max:0.111676 sd:0.0395482 dtype:DTYPE_FLOAT32 331: "embedding.330_of_512" NUMERICAL mean:0.00703036 min:-0.108652 max:0.103578 sd:0.0400975 dtype:DTYPE_FLOAT32 332: "embedding.331_of_512" NUMERICAL mean:0.000541923 min:-0.109862 max:0.10999 sd:0.0408574 dtype:DTYPE_FLOAT32 333: "embedding.332_of_512" NUMERICAL mean:0.0188891 min:-0.112872 max:0.118079 sd:0.0397373 dtype:DTYPE_FLOAT32 334: "embedding.333_of_512" NUMERICAL mean:-0.012192 min:-0.133506 max:0.13836 sd:0.0512842 dtype:DTYPE_FLOAT32 335: "embedding.334_of_512" NUMERICAL mean:-0.0265024 min:-0.126857 max:0.097852 sd:0.0420318 dtype:DTYPE_FLOAT32 336: "embedding.335_of_512" NUMERICAL mean:0.00215234 min:-0.111504 max:0.116062 sd:0.038159 dtype:DTYPE_FLOAT32 337: "embedding.336_of_512" NUMERICAL mean:-0.00825738 min:-0.125886 max:0.10212 sd:0.0376238 dtype:DTYPE_FLOAT32 338: "embedding.337_of_512" NUMERICAL mean:-0.0055194 min:-0.105159 max:0.110274 sd:0.0404973 dtype:DTYPE_FLOAT32 339: "embedding.338_of_512" NUMERICAL mean:0.0111058 min:-0.103003 max:0.134575 sd:0.0376746 dtype:DTYPE_FLOAT32 340: "embedding.339_of_512" NUMERICAL mean:0.00451027 min:-0.116598 max:0.114548 sd:0.0434438 dtype:DTYPE_FLOAT32 341: "embedding.340_of_512" NUMERICAL mean:0.0209382 min:-0.109457 max:0.119971 sd:0.0448743 dtype:DTYPE_FLOAT32 342: "embedding.341_of_512" NUMERICAL mean:0.00896807 min:-0.121829 max:0.10898 sd:0.0399955 dtype:DTYPE_FLOAT32 343: "embedding.342_of_512" NUMERICAL mean:-0.00661843 min:-0.113602 max:0.112046 sd:0.0417717 dtype:DTYPE_FLOAT32 344: "embedding.343_of_512" NUMERICAL mean:-0.00921778 min:-0.112399 max:0.116532 sd:0.0399069 dtype:DTYPE_FLOAT32 345: "embedding.344_of_512" NUMERICAL mean:0.00135801 min:-0.121002 max:0.0829257 sd:0.0322146 dtype:DTYPE_FLOAT32 346: "embedding.345_of_512" NUMERICAL mean:0.00347003 min:-0.131471 max:0.101491 sd:0.0404394 dtype:DTYPE_FLOAT32 347: "embedding.346_of_512" NUMERICAL mean:-0.00118125 min:-0.14804 max:0.11391 sd:0.0423848 dtype:DTYPE_FLOAT32 348: "embedding.347_of_512" NUMERICAL mean:-0.00893261 min:-0.125488 max:0.109213 sd:0.0498338 dtype:DTYPE_FLOAT32 349: "embedding.348_of_512" NUMERICAL mean:-0.0112279 min:-0.119783 max:0.106986 sd:0.039883 dtype:DTYPE_FLOAT32 350: "embedding.349_of_512" NUMERICAL mean:0.00921196 min:-0.108645 max:0.124486 sd:0.0427417 dtype:DTYPE_FLOAT32 351: "embedding.350_of_512" NUMERICAL mean:-0.0064119 min:-0.11853 max:0.108147 sd:0.0396107 dtype:DTYPE_FLOAT32 352: "embedding.351_of_512" NUMERICAL mean:0.00046816 min:-0.133059 max:0.106031 sd:0.0419676 dtype:DTYPE_FLOAT32 353: "embedding.352_of_512" NUMERICAL mean:0.00143986 min:-0.119083 max:0.0987319 sd:0.0358907 dtype:DTYPE_FLOAT32 354: "embedding.353_of_512" NUMERICAL mean:0.00247002 min:-0.109389 max:0.118887 sd:0.0416032 dtype:DTYPE_FLOAT32 355: "embedding.354_of_512" NUMERICAL mean:0.00010288 min:-0.139157 max:0.0995683 sd:0.0394998 dtype:DTYPE_FLOAT32 356: "embedding.355_of_512" NUMERICAL mean:-0.00525663 min:-0.146684 max:0.104288 sd:0.0406929 dtype:DTYPE_FLOAT32 357: "embedding.356_of_512" NUMERICAL mean:-0.0884722 min:-0.132905 max:0.0538598 sd:0.0108211 dtype:DTYPE_FLOAT32 358: "embedding.357_of_512" NUMERICAL mean:0.00677648 min:-0.110339 max:0.110136 sd:0.0402613 dtype:DTYPE_FLOAT32 359: "embedding.358_of_512" NUMERICAL mean:0.00630266 min:-0.111695 max:0.115859 sd:0.0427588 dtype:DTYPE_FLOAT32 360: "embedding.359_of_512" NUMERICAL mean:0.00225805 min:-0.126003 max:0.117678 sd:0.0444635 dtype:DTYPE_FLOAT32 361: "embedding.360_of_512" NUMERICAL mean:-0.00827234 min:-0.133543 max:0.115376 sd:0.0466799 dtype:DTYPE_FLOAT32 362: "embedding.361_of_512" NUMERICAL mean:-0.00625222 min:-0.10512 max:0.123856 sd:0.0418715 dtype:DTYPE_FLOAT32 363: "embedding.362_of_512" NUMERICAL mean:0.0651293 min:-0.11562 max:0.153915 sd:0.0359 dtype:DTYPE_FLOAT32 364: "embedding.363_of_512" NUMERICAL mean:0.00968887 min:-0.115793 max:0.11435 sd:0.0422501 dtype:DTYPE_FLOAT32 365: "embedding.364_of_512" NUMERICAL mean:0.00449241 min:-0.132071 max:0.103237 sd:0.0373741 dtype:DTYPE_FLOAT32 366: "embedding.365_of_512" NUMERICAL mean:-0.016221 min:-0.113495 max:0.106975 sd:0.0425689 dtype:DTYPE_FLOAT32 367: "embedding.366_of_512" NUMERICAL mean:0.0112515 min:-0.154925 max:0.151612 sd:0.0513015 dtype:DTYPE_FLOAT32 368: "embedding.367_of_512" NUMERICAL mean:-5.21381e-05 min:-0.11585 max:0.112307 sd:0.0391906 dtype:DTYPE_FLOAT32 369: "embedding.368_of_512" NUMERICAL mean:-0.00112394 min:-0.121213 max:0.126588 sd:0.044652 dtype:DTYPE_FLOAT32 370: "embedding.369_of_512" NUMERICAL mean:0.00485578 min:-0.106476 max:0.115632 sd:0.041426 dtype:DTYPE_FLOAT32 371: "embedding.370_of_512" NUMERICAL mean:-0.0174785 min:-0.114634 max:0.104434 sd:0.0382088 dtype:DTYPE_FLOAT32 372: "embedding.371_of_512" NUMERICAL mean:-0.00559737 min:-0.111149 max:0.115734 sd:0.0402863 dtype:DTYPE_FLOAT32 373: "embedding.372_of_512" NUMERICAL mean:-0.00348879 min:-0.108034 max:0.107825 sd:0.0403769 dtype:DTYPE_FLOAT32 374: "embedding.373_of_512" NUMERICAL mean:0.0188844 min:-0.127183 max:0.109232 sd:0.0404355 dtype:DTYPE_FLOAT32 375: "embedding.374_of_512" NUMERICAL mean:-0.00368461 min:-0.122589 max:0.124831 sd:0.0403308 dtype:DTYPE_FLOAT32 376: "embedding.375_of_512" NUMERICAL mean:-0.0106164 min:-0.118052 max:0.150001 sd:0.0432093 dtype:DTYPE_FLOAT32 377: "embedding.376_of_512" NUMERICAL mean:0.00311828 min:-0.106068 max:0.11577 sd:0.0400224 dtype:DTYPE_FLOAT32 378: "embedding.377_of_512" NUMERICAL mean:-0.0179061 min:-0.125819 max:0.111004 sd:0.0413477 dtype:DTYPE_FLOAT32 379: "embedding.378_of_512" NUMERICAL mean:-0.0129489 min:-0.126863 max:0.110993 sd:0.0434155 dtype:DTYPE_FLOAT32 380: "embedding.379_of_512" NUMERICAL mean:-0.00801255 min:-0.130591 max:0.112902 sd:0.04366 dtype:DTYPE_FLOAT32 381: "embedding.380_of_512" NUMERICAL mean:-0.00901065 min:-0.109901 max:0.123667 sd:0.0397827 dtype:DTYPE_FLOAT32 382: "embedding.381_of_512" NUMERICAL mean:0.00213499 min:-0.117992 max:0.104067 sd:0.0396603 dtype:DTYPE_FLOAT32 383: "embedding.382_of_512" NUMERICAL mean:0.0139051 min:-0.116796 max:0.115264 sd:0.041444 dtype:DTYPE_FLOAT32 384: "embedding.383_of_512" NUMERICAL mean:0.0015667 min:-0.137801 max:0.121558 sd:0.0446806 dtype:DTYPE_FLOAT32 385: "embedding.384_of_512" NUMERICAL mean:0.00590388 min:-0.136462 max:0.15641 sd:0.0551146 dtype:DTYPE_FLOAT32 386: "embedding.385_of_512" NUMERICAL mean:-0.0225046 min:-0.125096 max:0.122088 sd:0.0425471 dtype:DTYPE_FLOAT32 387: "embedding.386_of_512" NUMERICAL mean:-0.0291993 min:-0.149865 max:0.12312 sd:0.0469557 dtype:DTYPE_FLOAT32 388: "embedding.387_of_512" NUMERICAL mean:0.0136623 min:-0.113261 max:0.107316 sd:0.0408869 dtype:DTYPE_FLOAT32 389: "embedding.388_of_512" NUMERICAL mean:0.0119563 min:-0.0992984 max:0.118811 sd:0.0415827 dtype:DTYPE_FLOAT32 390: "embedding.389_of_512" NUMERICAL mean:1.88277e-05 min:-0.103729 max:0.117052 sd:0.0396793 dtype:DTYPE_FLOAT32 391: "embedding.390_of_512" NUMERICAL mean:0.00614745 min:-0.142472 max:0.132447 sd:0.0483092 dtype:DTYPE_FLOAT32 392: "embedding.391_of_512" NUMERICAL mean:-0.00252831 min:-0.111571 max:0.110414 sd:0.0407484 dtype:DTYPE_FLOAT32 393: "embedding.392_of_512" NUMERICAL mean:0.00560033 min:-0.106415 max:0.109868 sd:0.0411823 dtype:DTYPE_FLOAT32 394: "embedding.393_of_512" NUMERICAL mean:0.000437511 min:-0.115213 max:0.121544 sd:0.0406626 dtype:DTYPE_FLOAT32 395: "embedding.394_of_512" NUMERICAL mean:-0.00507897 min:-0.112722 max:0.112578 sd:0.0407212 dtype:DTYPE_FLOAT32 396: "embedding.395_of_512" NUMERICAL mean:-0.0104218 min:-0.106171 max:0.13395 sd:0.0412331 dtype:DTYPE_FLOAT32 397: "embedding.396_of_512" NUMERICAL mean:-0.025218 min:-0.121914 max:0.13782 sd:0.0420871 dtype:DTYPE_FLOAT32 398: "embedding.397_of_512" NUMERICAL mean:-0.00425221 min:-0.117618 max:0.106735 sd:0.0438757 dtype:DTYPE_FLOAT32 399: "embedding.398_of_512" NUMERICAL mean:-0.0112567 min:-0.136641 max:0.12107 sd:0.0385446 dtype:DTYPE_FLOAT32 400: "embedding.399_of_512" NUMERICAL mean:-0.00238481 min:-0.14689 max:0.132483 sd:0.0512686 dtype:DTYPE_FLOAT32 401: "embedding.400_of_512" NUMERICAL mean:-0.00854602 min:-0.110339 max:0.123831 sd:0.0428819 dtype:DTYPE_FLOAT32 402: "embedding.401_of_512" NUMERICAL mean:-0.0120933 min:-0.110716 max:0.107581 sd:0.0391564 dtype:DTYPE_FLOAT32 403: "embedding.402_of_512" NUMERICAL mean:-0.00798588 min:-0.114245 max:0.109355 sd:0.0417294 dtype:DTYPE_FLOAT32 404: "embedding.403_of_512" NUMERICAL mean:-0.00715776 min:-0.110958 max:0.109412 sd:0.0426725 dtype:DTYPE_FLOAT32 405: "embedding.404_of_512" NUMERICAL mean:0.0421547 min:-0.0936097 max:0.14341 sd:0.0449053 dtype:DTYPE_FLOAT32 406: "embedding.405_of_512" NUMERICAL mean:0.0138744 min:-0.101141 max:0.110993 sd:0.0409959 dtype:DTYPE_FLOAT32 407: "embedding.406_of_512" NUMERICAL mean:0.0221997 min:-0.10012 max:0.12351 sd:0.0512918 dtype:DTYPE_FLOAT32 408: "embedding.407_of_512" NUMERICAL mean:0.00840243 min:-0.100731 max:0.108785 sd:0.0385815 dtype:DTYPE_FLOAT32 409: "embedding.408_of_512" NUMERICAL mean:-0.00995255 min:-0.119931 max:0.107382 sd:0.0397331 dtype:DTYPE_FLOAT32 410: "embedding.409_of_512" NUMERICAL mean:0.00122281 min:-0.123687 max:0.110221 sd:0.0419264 dtype:DTYPE_FLOAT32 411: "embedding.410_of_512" NUMERICAL mean:0.00722765 min:-0.120324 max:0.118298 sd:0.0397953 dtype:DTYPE_FLOAT32 412: "embedding.411_of_512" NUMERICAL mean:0.00372596 min:-0.110838 max:0.104775 sd:0.0397102 dtype:DTYPE_FLOAT32 413: "embedding.412_of_512" NUMERICAL mean:0.00750692 min:-0.105861 max:0.113608 sd:0.0404272 dtype:DTYPE_FLOAT32 414: "embedding.413_of_512" NUMERICAL mean:0.00702045 min:-0.100497 max:0.109256 sd:0.040414 dtype:DTYPE_FLOAT32 415: "embedding.414_of_512" NUMERICAL mean:0.0129925 min:-0.104637 max:0.129069 sd:0.0476144 dtype:DTYPE_FLOAT32 416: "embedding.415_of_512" NUMERICAL mean:0.00895771 min:-0.103221 max:0.131867 sd:0.0416565 dtype:DTYPE_FLOAT32 417: "embedding.416_of_512" NUMERICAL mean:-0.0113754 min:-0.108457 max:0.108912 sd:0.039076 dtype:DTYPE_FLOAT32 418: "embedding.417_of_512" NUMERICAL mean:-0.00972072 min:-0.108896 max:0.120041 sd:0.039969 dtype:DTYPE_FLOAT32 419: "embedding.418_of_512" NUMERICAL mean:0.0103305 min:-0.115689 max:0.117791 sd:0.0438928 dtype:DTYPE_FLOAT32 420: "embedding.419_of_512" NUMERICAL mean:-0.011858 min:-0.110158 max:0.112286 sd:0.0405172 dtype:DTYPE_FLOAT32 421: "embedding.420_of_512" NUMERICAL mean:0.0113019 min:-0.117355 max:0.110719 sd:0.0390142 dtype:DTYPE_FLOAT32 422: "embedding.421_of_512" NUMERICAL mean:-0.00325833 min:-0.11971 max:0.0998387 sd:0.0386342 dtype:DTYPE_FLOAT32 423: "embedding.422_of_512" NUMERICAL mean:-0.0175019 min:-0.121014 max:0.108533 sd:0.0430717 dtype:DTYPE_FLOAT32 424: "embedding.423_of_512" NUMERICAL mean:0.00661466 min:-0.121052 max:0.104438 sd:0.0401472 dtype:DTYPE_FLOAT32 425: "embedding.424_of_512" NUMERICAL mean:0.0157025 min:-0.119043 max:0.121705 sd:0.0455012 dtype:DTYPE_FLOAT32 426: "embedding.425_of_512" NUMERICAL mean:0.00671776 min:-0.119955 max:0.135544 sd:0.046337 dtype:DTYPE_FLOAT32 427: "embedding.426_of_512" NUMERICAL mean:0.00625655 min:-0.110938 max:0.120801 sd:0.0434661 dtype:DTYPE_FLOAT32 428: "embedding.427_of_512" NUMERICAL mean:0.0204839 min:-0.112639 max:0.12859 sd:0.0461795 dtype:DTYPE_FLOAT32 429: "embedding.428_of_512" NUMERICAL mean:-0.00954845 min:-0.131481 max:0.103867 sd:0.0409481 dtype:DTYPE_FLOAT32 430: "embedding.429_of_512" NUMERICAL mean:0.0227497 min:-0.114759 max:0.128784 sd:0.0461912 dtype:DTYPE_FLOAT32 431: "embedding.430_of_512" NUMERICAL mean:-0.0143054 min:-0.116372 max:0.0982788 sd:0.0397653 dtype:DTYPE_FLOAT32 432: "embedding.431_of_512" NUMERICAL mean:0.00108119 min:-0.10975 max:0.113431 sd:0.0395805 dtype:DTYPE_FLOAT32 433: "embedding.432_of_512" NUMERICAL mean:-0.0124634 min:-0.128303 max:0.122121 sd:0.043612 dtype:DTYPE_FLOAT32 434: "embedding.433_of_512" NUMERICAL mean:-0.000974066 min:-0.127452 max:0.143975 sd:0.0512878 dtype:DTYPE_FLOAT32 435: "embedding.434_of_512" NUMERICAL mean:-0.000695708 min:-0.117519 max:0.132419 sd:0.048299 dtype:DTYPE_FLOAT32 436: "embedding.435_of_512" NUMERICAL mean:-0.00800422 min:-0.11716 max:0.106095 sd:0.0385783 dtype:DTYPE_FLOAT32 437: "embedding.436_of_512" NUMERICAL mean:-0.00449899 min:-0.119801 max:0.13136 sd:0.0450766 dtype:DTYPE_FLOAT32 438: "embedding.437_of_512" NUMERICAL mean:0.00152719 min:-0.101368 max:0.111586 sd:0.0373092 dtype:DTYPE_FLOAT32 439: "embedding.438_of_512" NUMERICAL mean:-0.00746199 min:-0.110446 max:0.107505 sd:0.0409118 dtype:DTYPE_FLOAT32 440: "embedding.439_of_512" NUMERICAL mean:-0.000542517 min:-0.126726 max:0.150725 sd:0.0498822 dtype:DTYPE_FLOAT32 441: "embedding.440_of_512" NUMERICAL mean:0.0162618 min:-0.110413 max:0.112766 sd:0.039636 dtype:DTYPE_FLOAT32 442: "embedding.441_of_512" NUMERICAL mean:-0.0252852 min:-0.140847 max:0.123998 sd:0.045552 dtype:DTYPE_FLOAT32 443: "embedding.442_of_512" NUMERICAL mean:-0.00971423 min:-0.14093 max:0.115633 sd:0.0430468 dtype:DTYPE_FLOAT32 444: "embedding.443_of_512" NUMERICAL mean:-0.00171618 min:-0.130186 max:0.122902 sd:0.0446095 dtype:DTYPE_FLOAT32 445: "embedding.444_of_512" NUMERICAL mean:0.0108986 min:-0.114492 max:0.110956 sd:0.0418642 dtype:DTYPE_FLOAT32 446: "embedding.445_of_512" NUMERICAL mean:-0.00650931 min:-0.106713 max:0.126819 sd:0.0394136 dtype:DTYPE_FLOAT32 447: "embedding.446_of_512" NUMERICAL mean:7.68803e-05 min:-0.107121 max:0.104196 sd:0.0371536 dtype:DTYPE_FLOAT32 448: "embedding.447_of_512" NUMERICAL mean:-0.00166973 min:-0.106304 max:0.113193 sd:0.0417721 dtype:DTYPE_FLOAT32 449: "embedding.448_of_512" NUMERICAL mean:0.00143107 min:-0.112879 max:0.117707 sd:0.0438514 dtype:DTYPE_FLOAT32 450: "embedding.449_of_512" NUMERICAL mean:0.00577755 min:-0.114301 max:0.116267 sd:0.0413021 dtype:DTYPE_FLOAT32 451: "embedding.450_of_512" NUMERICAL mean:0.00523777 min:-0.121324 max:0.109753 sd:0.0422962 dtype:DTYPE_FLOAT32 452: "embedding.451_of_512" NUMERICAL mean:0.00232381 min:-0.107421 max:0.116006 sd:0.0411045 dtype:DTYPE_FLOAT32 453: "embedding.452_of_512" NUMERICAL mean:0.0131371 min:-0.119915 max:0.110052 sd:0.0388742 dtype:DTYPE_FLOAT32 454: "embedding.453_of_512" NUMERICAL mean:0.00384022 min:-0.113448 max:0.103866 sd:0.0399839 dtype:DTYPE_FLOAT32 455: "embedding.454_of_512" NUMERICAL mean:0.00746132 min:-0.11867 max:0.107229 sd:0.0393659 dtype:DTYPE_FLOAT32 456: "embedding.455_of_512" NUMERICAL mean:0.0217711 min:-0.130108 max:0.130266 sd:0.0457751 dtype:DTYPE_FLOAT32 457: "embedding.456_of_512" NUMERICAL mean:-0.00486573 min:-0.125269 max:0.103216 sd:0.0417326 dtype:DTYPE_FLOAT32 458: "embedding.457_of_512" NUMERICAL mean:-0.00370284 min:-0.152411 max:0.118391 sd:0.0475716 dtype:DTYPE_FLOAT32 459: "embedding.458_of_512" NUMERICAL mean:-0.0252088 min:-0.129244 max:0.110772 sd:0.0447074 dtype:DTYPE_FLOAT32 460: "embedding.459_of_512" NUMERICAL mean:0.0196455 min:-0.107007 max:0.110025 sd:0.0371162 dtype:DTYPE_FLOAT32 461: "embedding.460_of_512" NUMERICAL mean:0.00486969 min:-0.133702 max:0.117438 sd:0.0404765 dtype:DTYPE_FLOAT32 462: "embedding.461_of_512" NUMERICAL mean:0.00879324 min:-0.123721 max:0.109769 sd:0.0418885 dtype:DTYPE_FLOAT32 463: "embedding.462_of_512" NUMERICAL mean:0.00541842 min:-0.103881 max:0.115937 sd:0.040526 dtype:DTYPE_FLOAT32 464: "embedding.463_of_512" NUMERICAL mean:0.0112013 min:-0.129965 max:0.125135 sd:0.0445652 dtype:DTYPE_FLOAT32 465: "embedding.464_of_512" NUMERICAL mean:-0.00978469 min:-0.112536 max:0.136367 sd:0.0432779 dtype:DTYPE_FLOAT32 466: "embedding.465_of_512" NUMERICAL mean:-0.00372292 min:-0.132975 max:0.107404 sd:0.0434915 dtype:DTYPE_FLOAT32 467: "embedding.466_of_512" NUMERICAL mean:0.000832962 min:-0.106678 max:0.109534 sd:0.041454 dtype:DTYPE_FLOAT32 468: "embedding.467_of_512" NUMERICAL mean:0.0128707 min:-0.123202 max:0.108301 sd:0.036966 dtype:DTYPE_FLOAT32 469: "embedding.468_of_512" NUMERICAL mean:0.00143448 min:-0.109754 max:0.115596 sd:0.0410802 dtype:DTYPE_FLOAT32 470: "embedding.469_of_512" NUMERICAL mean:0.00821259 min:-0.0968573 max:0.116681 sd:0.037229 dtype:DTYPE_FLOAT32 471: "embedding.470_of_512" NUMERICAL mean:-0.0126405 min:-0.11236 max:0.104975 sd:0.0410705 dtype:DTYPE_FLOAT32 472: "embedding.471_of_512" NUMERICAL mean:0.00967789 min:-0.114741 max:0.113365 sd:0.0415494 dtype:DTYPE_FLOAT32 473: "embedding.472_of_512" NUMERICAL mean:0.0051147 min:-0.116287 max:0.123708 sd:0.038196 dtype:DTYPE_FLOAT32 474: "embedding.473_of_512" NUMERICAL mean:0.00460656 min:-0.117806 max:0.116034 sd:0.0417151 dtype:DTYPE_FLOAT32 475: "embedding.474_of_512" NUMERICAL mean:-0.00244138 min:-0.103319 max:0.116585 sd:0.0374234 dtype:DTYPE_FLOAT32 476: "embedding.475_of_512" NUMERICAL mean:-0.00797766 min:-0.112168 max:0.110854 sd:0.043268 dtype:DTYPE_FLOAT32 477: "embedding.476_of_512" NUMERICAL mean:-0.0123356 min:-0.118527 max:0.110389 sd:0.0415487 dtype:DTYPE_FLOAT32 478: "embedding.477_of_512" NUMERICAL mean:-0.00891097 min:-0.109911 max:0.114824 sd:0.0409558 dtype:DTYPE_FLOAT32 479: "embedding.478_of_512" NUMERICAL mean:0.0531792 min:-0.123494 max:0.14429 sd:0.0446525 dtype:DTYPE_FLOAT32 480: "embedding.479_of_512" NUMERICAL mean:0.00310177 min:-0.126525 max:0.135642 sd:0.0508086 dtype:DTYPE_FLOAT32 481: "embedding.480_of_512" NUMERICAL mean:0.00178777 min:-0.101512 max:0.111535 sd:0.0393616 dtype:DTYPE_FLOAT32 482: "embedding.481_of_512" NUMERICAL mean:0.000436977 min:-0.141595 max:0.116526 sd:0.0498771 dtype:DTYPE_FLOAT32 483: "embedding.482_of_512" NUMERICAL mean:0.0139387 min:-0.109079 max:0.125151 sd:0.0395955 dtype:DTYPE_FLOAT32 484: "embedding.483_of_512" NUMERICAL mean:-0.0190178 min:-0.116579 max:0.12211 sd:0.0404221 dtype:DTYPE_FLOAT32 485: "embedding.484_of_512" NUMERICAL mean:0.0111983 min:-0.115318 max:0.114151 sd:0.0407415 dtype:DTYPE_FLOAT32 486: "embedding.485_of_512" NUMERICAL mean:-0.0210413 min:-0.12817 max:0.102505 sd:0.0409111 dtype:DTYPE_FLOAT32 487: "embedding.486_of_512" NUMERICAL mean:0.00291598 min:-0.136717 max:0.132649 sd:0.0483605 dtype:DTYPE_FLOAT32 488: "embedding.487_of_512" NUMERICAL mean:0.0258506 min:-0.118507 max:0.139141 sd:0.0476916 dtype:DTYPE_FLOAT32 489: "embedding.488_of_512" NUMERICAL mean:0.00950834 min:-0.117085 max:0.104573 sd:0.0394689 dtype:DTYPE_FLOAT32 490: "embedding.489_of_512" NUMERICAL mean:-0.00655678 min:-0.113501 max:0.116317 sd:0.0412641 dtype:DTYPE_FLOAT32 491: "embedding.490_of_512" NUMERICAL mean:0.0025444 min:-0.0976397 max:0.133059 sd:0.0391231 dtype:DTYPE_FLOAT32 492: "embedding.491_of_512" NUMERICAL mean:-0.00116524 min:-0.115012 max:0.108975 sd:0.0373331 dtype:DTYPE_FLOAT32 493: "embedding.492_of_512" NUMERICAL mean:-0.00805514 min:-0.112223 max:0.118394 sd:0.0409569 dtype:DTYPE_FLOAT32 494: "embedding.493_of_512" NUMERICAL mean:-0.00381922 min:-0.109779 max:0.113538 sd:0.0375221 dtype:DTYPE_FLOAT32 495: "embedding.494_of_512" NUMERICAL mean:0.0192517 min:-0.108658 max:0.118238 sd:0.0414103 dtype:DTYPE_FLOAT32 496: "embedding.495_of_512" NUMERICAL mean:-0.00252727 min:-0.118617 max:0.100404 sd:0.0398346 dtype:DTYPE_FLOAT32 497: "embedding.496_of_512" NUMERICAL mean:-0.000870086 min:-0.10941 max:0.119059 sd:0.043479 dtype:DTYPE_FLOAT32 498: "embedding.497_of_512" NUMERICAL mean:-0.00296294 min:-0.123757 max:0.109776 sd:0.0420959 dtype:DTYPE_FLOAT32 499: "embedding.498_of_512" NUMERICAL mean:0.0127804 min:-0.138546 max:0.154906 sd:0.0511673 dtype:DTYPE_FLOAT32 500: "embedding.499_of_512" NUMERICAL mean:-0.00481274 min:-0.104637 max:0.112387 sd:0.0419786 dtype:DTYPE_FLOAT32 501: "embedding.500_of_512" NUMERICAL mean:-0.012455 min:-0.109502 max:0.102241 sd:0.0402451 dtype:DTYPE_FLOAT32 502: "embedding.501_of_512" NUMERICAL mean:-0.0236005 min:-0.117228 max:0.124977 sd:0.0464432 dtype:DTYPE_FLOAT32 503: "embedding.502_of_512" NUMERICAL mean:0.00916425 min:-0.128705 max:0.110148 sd:0.0412428 dtype:DTYPE_FLOAT32 504: "embedding.503_of_512" NUMERICAL mean:-0.0099854 min:-0.179229 max:0.112813 sd:0.0666002 dtype:DTYPE_FLOAT32 505: "embedding.504_of_512" NUMERICAL mean:0.0140659 min:-0.124558 max:0.131239 sd:0.0459631 dtype:DTYPE_FLOAT32 506: "embedding.505_of_512" NUMERICAL mean:0.00529723 min:-0.119894 max:0.104362 sd:0.0399805 dtype:DTYPE_FLOAT32 507: "embedding.506_of_512" NUMERICAL mean:-0.00319069 min:-0.111178 max:0.108562 sd:0.040611 dtype:DTYPE_FLOAT32 508: "embedding.507_of_512" NUMERICAL mean:-0.00332249 min:-0.108088 max:0.118358 sd:0.0396039 dtype:DTYPE_FLOAT32 509: "embedding.508_of_512" NUMERICAL mean:-0.00396023 min:-0.11048 max:0.107852 sd:0.0375341 dtype:DTYPE_FLOAT32 510: "embedding.509_of_512" NUMERICAL mean:-0.00917504 min:-0.116661 max:0.100524 sd:0.0361387 dtype:DTYPE_FLOAT32 511: "embedding.510_of_512" NUMERICAL mean:0.037723 min:-0.0965471 max:0.140981 sd:0.0479428 dtype:DTYPE_FLOAT32 512: "embedding.511_of_512" NUMERICAL mean:0.00788656 min:-0.116457 max:0.102988 sd:0.0402552 dtype:DTYPE_FLOAT32 CATEGORICAL: 1 (0.194932%) 0: "label" CATEGORICAL has-dict vocab-size:3 zero-ood-items most-frequent:"1" 37569 (55.7826%) dtype:DTYPE_INT64 Terminology: nas: Number of non-available (i.e. missing) values. ood: Out of dictionary. manually-defined: Attribute whose type is manually defined by the user, i.e., the type was not automatically inferred. tokenized: The attribute value is obtained through tokenization. has-dict: The attribute is attached to a string dictionary e.g. a categorical attribute stored as a string. vocab-size: Number of unique values.
The following evaluation is computed on the validation or out-of-bag dataset.
Task: CLASSIFICATION
Label: label
Loss (BINOMIAL_LOG_LIKELIHOOD): 0.738938
Accuracy: 0.837441 CI95[W][0 1]
ErrorRate: : 0.162559
Confusion Table:
truth\prediction
1 0
1 3230 559
0 536 2411
Total: 6736
Variable importances measure the importance of an input feature for a model.
1. "embedding.171_of_512" 0.209350 ################
2. "embedding.111_of_512" 0.204045 #############
3. "embedding.384_of_512" 0.178102 ####
4. "embedding.291_of_512" 0.177804 ###
5. "embedding.408_of_512" 0.177389 ###
6. "embedding.365_of_512" 0.176330 ###
7. "embedding.188_of_512" 0.175333 ###
8. "embedding.276_of_512" 0.175256 ##
9. "embedding.036_of_512" 0.174208 ##
10. "embedding.350_of_512" 0.174185 ##
11. "embedding.343_of_512" 0.174146 ##
12. "embedding.022_of_512" 0.173972 ##
13. "embedding.333_of_512" 0.173661 ##
14. "embedding.332_of_512" 0.173290 ##
15. "embedding.307_of_512" 0.172651 ##
16. "embedding.058_of_512" 0.172588 #
17. "embedding.167_of_512" 0.172248 #
18. "embedding.184_of_512" 0.172189 #
19. "embedding.041_of_512" 0.172093 #
20. "embedding.118_of_512" 0.171974 #
21. "embedding.063_of_512" 0.171966 #
22. "embedding.065_of_512" 0.171925 #
23. "embedding.132_of_512" 0.171918 #
24. "embedding.346_of_512" 0.171654 #
25. "embedding.092_of_512" 0.171654 #
26. "embedding.182_of_512" 0.171428 #
27. "embedding.043_of_512" 0.171327 #
28. "embedding.329_of_512" 0.171311 #
29. "embedding.342_of_512" 0.171305 #
30. "embedding.140_of_512" 0.171297 #
31. "embedding.302_of_512" 0.171015 #
32. "embedding.297_of_512" 0.170923 #
33. "embedding.478_of_512" 0.170846 #
34. "embedding.511_of_512" 0.170711 #
35. "embedding.227_of_512" 0.170582 #
36. "embedding.175_of_512" 0.170326 #
37. "embedding.237_of_512" 0.170321 #
38. "embedding.292_of_512" 0.170164 #
39. "embedding.363_of_512" 0.170138 #
40. "embedding.319_of_512" 0.169955
41. "embedding.424_of_512" 0.169759
42. "embedding.079_of_512" 0.169743
43. "embedding.306_of_512" 0.169673
44. "embedding.368_of_512" 0.169648
45. "embedding.494_of_512" 0.169495
46. "embedding.052_of_512" 0.169474
47. "embedding.194_of_512" 0.169375
48. "embedding.483_of_512" 0.169328
49. "embedding.240_of_512" 0.169305
50. "embedding.046_of_512" 0.169244
51. "embedding.243_of_512" 0.169234
52. "embedding.215_of_512" 0.169210
53. "embedding.152_of_512" 0.169186
54. "embedding.502_of_512" 0.169168
55. "embedding.471_of_512" 0.169095
56. "embedding.376_of_512" 0.169079
57. "embedding.244_of_512" 0.169073
58. "embedding.219_of_512" 0.169046
59. "embedding.308_of_512" 0.169035
60. "embedding.287_of_512" 0.169025
61. "embedding.044_of_512" 0.169018
62. "embedding.195_of_512" 0.168966
63. "embedding.130_of_512" 0.168943
64. "embedding.469_of_512" 0.168916
65. "embedding.352_of_512" 0.168860
66. "embedding.358_of_512" 0.168809
67. "embedding.289_of_512" 0.168801
68. "embedding.142_of_512" 0.168721
69. "embedding.503_of_512" 0.168673
70. "embedding.205_of_512" 0.168658
71. "embedding.126_of_512" 0.168630
72. "embedding.116_of_512" 0.168582
73. "embedding.185_of_512" 0.168548
74. "embedding.228_of_512" 0.168537
75. "embedding.225_of_512" 0.168512
76. "embedding.427_of_512" 0.168510
77. "embedding.443_of_512" 0.168502
78. "embedding.267_of_512" 0.168497
79. "embedding.399_of_512" 0.168497
80. "embedding.050_of_512" 0.168446
81. "embedding.158_of_512" 0.168424
82. "embedding.449_of_512" 0.168423
83. "embedding.202_of_512" 0.168404
84. "embedding.066_of_512" 0.168397
85. "embedding.155_of_512" 0.168383
86. "embedding.012_of_512" 0.168377
87. "embedding.268_of_512" 0.168339
88. "embedding.327_of_512" 0.168327
89. "embedding.359_of_512" 0.168288
90. "embedding.072_of_512" 0.168259
91. "embedding.416_of_512" 0.168239
92. "embedding.093_of_512" 0.168229
93. "embedding.169_of_512" 0.168224
94. "embedding.433_of_512" 0.168222
95. "embedding.233_of_512" 0.168221
96. "embedding.068_of_512" 0.168220
97. "embedding.098_of_512" 0.168219
98. "embedding.415_of_512" 0.168218
99. "embedding.076_of_512" 0.168206
100. "embedding.200_of_512" 0.168200
101. "embedding.314_of_512" 0.168172
102. "embedding.197_of_512" 0.168162
103. "embedding.030_of_512" 0.168141
104. "embedding.141_of_512" 0.168119
105. "embedding.498_of_512" 0.168115
106. "embedding.402_of_512" 0.168106
107. "embedding.321_of_512" 0.168090
108. "embedding.479_of_512" 0.168059
109. "embedding.146_of_512" 0.168056
110. "embedding.370_of_512" 0.168052
111. "embedding.051_of_512" 0.168045
112. "embedding.262_of_512" 0.168041
113. "embedding.163_of_512" 0.168039
114. "embedding.187_of_512" 0.168037
115. "embedding.247_of_512" 0.168030
116. "embedding.390_of_512" 0.168029
117. "embedding.204_of_512" 0.168024
118. "embedding.179_of_512" 0.168023
119. "embedding.235_of_512" 0.168017
120. "embedding.378_of_512" 0.168014
121. "embedding.209_of_512" 0.168010
122. "embedding.057_of_512" 0.168008
123. "embedding.505_of_512" 0.168005
124. "embedding.286_of_512" 0.168004
125. "embedding.296_of_512" 0.167997
126. "embedding.313_of_512" 0.167978
127. "embedding.356_of_512" 0.167977
128. "embedding.437_of_512" 0.167975
129. "embedding.489_of_512" 0.167952
130. "embedding.338_of_512" 0.167946
131. "embedding.283_of_512" 0.167939
132. "embedding.071_of_512" 0.167930
133. "embedding.173_of_512" 0.167921
134. "embedding.405_of_512" 0.167918
135. "embedding.114_of_512" 0.167901
136. "embedding.136_of_512" 0.167900
137. "embedding.281_of_512" 0.167900
138. "embedding.064_of_512" 0.167894
139. "embedding.284_of_512" 0.167878
140. "embedding.060_of_512" 0.167857
141. "embedding.061_of_512" 0.167852
142. "embedding.375_of_512" 0.167851
143. "embedding.039_of_512" 0.167847
144. "embedding.042_of_512" 0.167837
145. "embedding.293_of_512" 0.167836
146. "embedding.218_of_512" 0.167832
147. "embedding.255_of_512" 0.167831
148. "embedding.067_of_512" 0.167826
149. "embedding.326_of_512" 0.167825
150. "embedding.110_of_512" 0.167818
151. "embedding.191_of_512" 0.167814
152. "embedding.263_of_512" 0.167808
153. "embedding.020_of_512" 0.167807
154. "embedding.115_of_512" 0.167807
155. "embedding.265_of_512" 0.167801
156. "embedding.324_of_512" 0.167796
157. "embedding.011_of_512" 0.167792
158. "embedding.090_of_512" 0.167789
159. "embedding.180_of_512" 0.167769
160. "embedding.381_of_512" 0.167767
161. "embedding.362_of_512" 0.167762
162. "embedding.054_of_512" 0.167759
163. "embedding.382_of_512" 0.167751
164. "embedding.325_of_512" 0.167749
165. "embedding.454_of_512" 0.167739
166. "embedding.251_of_512" 0.167737
167. "embedding.331_of_512" 0.167733
168. "embedding.480_of_512" 0.167732
169. "embedding.221_of_512" 0.167730
170. "embedding.217_of_512" 0.167729
171. "embedding.354_of_512" 0.167722
172. "embedding.391_of_512" 0.167720
173. "embedding.444_of_512" 0.167708
174. "embedding.216_of_512" 0.167708
175. "embedding.232_of_512" 0.167694
176. "embedding.143_of_512" 0.167693
177. "embedding.119_of_512" 0.167688
178. "embedding.123_of_512" 0.167683
179. "embedding.095_of_512" 0.167677
180. "embedding.018_of_512" 0.167676
181. "embedding.144_of_512" 0.167676
182. "embedding.062_of_512" 0.167672
183. "embedding.021_of_512" 0.167666
184. "embedding.165_of_512" 0.167664
185. "embedding.208_of_512" 0.167661
186. "embedding.078_of_512" 0.167659
187. "embedding.309_of_512" 0.167658
188. "embedding.113_of_512" 0.167654
189. "embedding.080_of_512" 0.167653
190. "embedding.386_of_512" 0.167653
191. "embedding.447_of_512" 0.167653
192. "embedding.315_of_512" 0.167653
193. "embedding.279_of_512" 0.167645
194. "embedding.135_of_512" 0.167641
195. "embedding.411_of_512" 0.167640
196. "embedding.107_of_512" 0.167640
197. "embedding.082_of_512" 0.167638
198. "embedding.421_of_512" 0.167637
199. "embedding.089_of_512" 0.167637
200. "embedding.059_of_512" 0.167636
201. "embedding.317_of_512" 0.167636
202. "embedding.410_of_512" 0.167636
203. "embedding.305_of_512" 0.167634
204. "embedding.290_of_512" 0.167631
205. "embedding.056_of_512" 0.167630
206. "embedding.190_of_512" 0.167629
207. "embedding.094_of_512" 0.167628
208. "embedding.400_of_512" 0.167625
209. "embedding.439_of_512" 0.167624
210. "embedding.193_of_512" 0.167623
211. "embedding.406_of_512" 0.167619
212. "embedding.423_of_512" 0.167617
213. "embedding.212_of_512" 0.167607
214. "embedding.003_of_512" 0.167601
215. "embedding.105_of_512" 0.167595
216. "embedding.256_of_512" 0.167589
217. "embedding.086_of_512" 0.167586
218. "embedding.417_of_512" 0.167586
219. "embedding.482_of_512" 0.167585
220. "embedding.192_of_512" 0.167585
221. "embedding.040_of_512" 0.167584
222. "embedding.236_of_512" 0.167584
223. "embedding.178_of_512" 0.167583
224. "embedding.183_of_512" 0.167580
225. "embedding.025_of_512" 0.167579
226. "embedding.069_of_512" 0.167577
227. "embedding.379_of_512" 0.167571
228. "embedding.081_of_512" 0.167569
229. "embedding.431_of_512" 0.167569
230. "embedding.492_of_512" 0.167566
231. "embedding.371_of_512" 0.167561
232. "embedding.451_of_512" 0.167561
233. "embedding.456_of_512" 0.167558
234. "embedding.322_of_512" 0.167558
235. "embedding.024_of_512" 0.167555
236. "embedding.345_of_512" 0.167554
237. "embedding.426_of_512" 0.167554
238. "embedding.495_of_512" 0.167554
239. "embedding.463_of_512" 0.167553
240. "embedding.122_of_512" 0.167551
241. "embedding.477_of_512" 0.167551
242. "embedding.250_of_512" 0.167551
243. "embedding.074_of_512" 0.167549
244. "embedding.150_of_512" 0.167548
245. "embedding.261_of_512" 0.167548
246. "embedding.133_of_512" 0.167547
247. "embedding.389_of_512" 0.167546
248. "embedding.070_of_512" 0.167543
249. "embedding.245_of_512" 0.167541
250. "embedding.303_of_512" 0.167541
251. "embedding.344_of_512" 0.167540
252. "embedding.499_of_512" 0.167534
253. "embedding.418_of_512" 0.167534
254. "embedding.462_of_512" 0.167534
255. "embedding.470_of_512" 0.167533
256. "embedding.087_of_512" 0.167533
257. "embedding.038_of_512" 0.167533
258. "embedding.096_of_512" 0.167533
259. "embedding.223_of_512" 0.167533
260. "embedding.438_of_512" 0.167531
261. "embedding.026_of_512" 0.167530
262. "embedding.153_of_512" 0.167524
263. "embedding.347_of_512" 0.167520
264. "embedding.468_of_512" 0.167520
265. "embedding.474_of_512" 0.167519
266. "embedding.366_of_512" 0.167516
267. "embedding.484_of_512" 0.167516
268. "embedding.206_of_512" 0.167514
269. "embedding.101_of_512" 0.167513
270. "embedding.288_of_512" 0.167513
271. "embedding.373_of_512" 0.167513
272. "embedding.465_of_512" 0.167513
273. "embedding.341_of_512" 0.167507
274. "embedding.014_of_512" 0.167506
275. "embedding.160_of_512" 0.167503
276. "embedding.493_of_512" 0.167502
277. "embedding.264_of_512" 0.167501
278. "embedding.508_of_512" 0.167500
279. "embedding.000_of_512" 0.167498
280. "embedding.161_of_512" 0.167498
281. "embedding.189_of_512" 0.167498
282. "embedding.320_of_512" 0.167498
283. "embedding.159_of_512" 0.167496
284. "embedding.458_of_512" 0.167496
285. "embedding.075_of_512" 0.167496
286. "embedding.496_of_512" 0.167496
287. "embedding.385_of_512" 0.167495
288. "embedding.023_of_512" 0.167488
289. "embedding.361_of_512" 0.167485
290. "embedding.388_of_512" 0.167485
291. "embedding.137_of_512" 0.167484
292. "embedding.337_of_512" 0.167483
293. "embedding.487_of_512" 0.167483
294. "embedding.053_of_512" 0.167482
295. "embedding.164_of_512" 0.167482
296. "embedding.230_of_512" 0.167481
297. "embedding.168_of_512" 0.167481
298. "embedding.138_of_512" 0.167481
299. "embedding.131_of_512" 0.167481
300. "embedding.269_of_512" 0.167481
301. "embedding.340_of_512" 0.167480
302. "embedding.174_of_512" 0.167480
303. "embedding.372_of_512" 0.167480
304. "embedding.177_of_512" 0.167480
305. "embedding.472_of_512" 0.167479
306. "embedding.422_of_512" 0.167479
307. "embedding.461_of_512" 0.167479
308. "embedding.220_of_512" 0.167478
309. "embedding.348_of_512" 0.167478
310. "embedding.100_of_512" 0.167478
311. "embedding.005_of_512" 0.167478
312. "embedding.104_of_512" 0.167478
313. "embedding.239_of_512" 0.167478
314. "embedding.266_of_512" 0.167478
315. "embedding.336_of_512" 0.167478
316. "embedding.510_of_512" 0.167478
317. "embedding.045_of_512" 0.167473
318. "embedding.156_of_512" 0.167473
319. "embedding.234_of_512" 0.167473
320. "embedding.277_of_512" 0.167473
321. "embedding.009_of_512" 0.167470
322. "embedding.412_of_512" 0.167470
323. "embedding.304_of_512" 0.167466
324. "embedding.316_of_512" 0.167466
325. "embedding.035_of_512" 0.167464
326. "embedding.460_of_512" 0.167464
327. "embedding.091_of_512" 0.167463
328. "embedding.278_of_512" 0.167463
329. "embedding.272_of_512" 0.167463
330. "embedding.441_of_512" 0.167463
331. "embedding.033_of_512" 0.167461
332. "embedding.445_of_512" 0.167461
333. "embedding.016_of_512" 0.167461
334. "embedding.259_of_512" 0.167461
335. "embedding.357_of_512" 0.167461
336. "embedding.049_of_512" 0.167461
337. "embedding.134_of_512" 0.167461
338. "embedding.207_of_512" 0.167461
339. "embedding.213_of_512" 0.167461
340. "embedding.008_of_512" 0.167460
341. "embedding.258_of_512" 0.167460
342. "embedding.328_of_512" 0.167460
343. "embedding.353_of_512" 0.167460
344. "embedding.501_of_512" 0.167460
345. "embedding.383_of_512" 0.167456
346. "embedding.004_of_512" 0.167455
347. "embedding.224_of_512" 0.167447
348. "embedding.323_of_512" 0.167446
349. "embedding.294_of_512" 0.167445
350. "embedding.084_of_512" 0.167445
351. "embedding.475_of_512" 0.167445
352. "embedding.015_of_512" 0.167445
353. "embedding.210_of_512" 0.167444
354. "embedding.109_of_512" 0.167444
355. "embedding.442_of_512" 0.167444
356. "embedding.027_of_512" 0.167443
357. "embedding.300_of_512" 0.167443
358. "embedding.369_of_512" 0.167443
359. "embedding.394_of_512" 0.167443
360. "embedding.409_of_512" 0.167443
361. "embedding.485_of_512" 0.167443
362. "embedding.031_of_512" 0.167443
363. "embedding.139_of_512" 0.167437
364. "embedding.199_of_512" 0.167437
365. "embedding.226_of_512" 0.167437
366. "embedding.001_of_512" 0.167430
367. "embedding.428_of_512" 0.167430
368. "embedding.440_of_512" 0.167428
369. "embedding.452_of_512" 0.167427
370. "embedding.260_of_512" 0.167427
371. "embedding.238_of_512" 0.167427
372. "embedding.351_of_512" 0.167427
373. "embedding.097_of_512" 0.167426
374. "embedding.125_of_512" 0.167426
375. "embedding.403_of_512" 0.167426
376. "embedding.006_of_512" 0.167426
377. "embedding.312_of_512" 0.167426
378. "embedding.010_of_512" 0.167426
379. "embedding.280_of_512" 0.167426
380. "embedding.466_of_512" 0.167426
381. "embedding.151_of_512" 0.167425
382. "embedding.476_of_512" 0.167425
383. "embedding.085_of_512" 0.167425
384. "embedding.154_of_512" 0.167419
385. "embedding.310_of_512" 0.167419
386. "embedding.413_of_512" 0.167418
387. "embedding.274_of_512" 0.167415
388. "embedding.007_of_512" 0.167412
389. "embedding.120_of_512" 0.167412
390. "embedding.231_of_512" 0.167411
391. "embedding.360_of_512" 0.167411
392. "embedding.380_of_512" 0.167411
393. "embedding.106_of_512" 0.167410
394. "embedding.170_of_512" 0.167410
395. "embedding.172_of_512" 0.167410
396. "embedding.257_of_512" 0.167410
397. "embedding.077_of_512" 0.167409
398. "embedding.157_of_512" 0.167409
399. "embedding.434_of_512" 0.167409
400. "embedding.032_of_512" 0.167409
401. "embedding.083_of_512" 0.167409
402. "embedding.198_of_512" 0.167408
403. "embedding.229_of_512" 0.167408
404. "embedding.271_of_512" 0.167408
405. "embedding.430_of_512" 0.167408
406. "embedding.432_of_512" 0.167408
407. "embedding.467_of_512" 0.167408
408. "embedding.500_of_512" 0.167408
409. "embedding.507_of_512" 0.167408
410. "embedding.028_of_512" 0.167408
411. "embedding.037_of_512" 0.167408
412. "embedding.145_of_512" 0.167408
413. "embedding.301_of_512" 0.167408
414. "embedding.457_of_512" 0.167408
415. "embedding.464_of_512" 0.167408
416. "embedding.102_of_512" 0.167408
417. "embedding.330_of_512" 0.167408
418. "embedding.404_of_512" 0.167408
419. "embedding.425_of_512" 0.167408
420. "embedding.002_of_512" 0.167408
421. "embedding.055_of_512" 0.167408
422. "embedding.073_of_512" 0.167408
423. "embedding.124_of_512" 0.167408
424. "embedding.211_of_512" 0.167408
425. "embedding.364_of_512" 0.167408
426. "embedding.459_of_512" 0.167408
427. "embedding.335_of_512" 0.167394
428. "embedding.034_of_512" 0.167394
429. "embedding.401_of_512" 0.167393
430. "embedding.214_of_512" 0.167392
431. "embedding.252_of_512" 0.167392
432. "embedding.481_of_512" 0.167392
433. "embedding.029_of_512" 0.167391
434. "embedding.149_of_512" 0.167391
435. "embedding.196_of_512" 0.167391
436. "embedding.282_of_512" 0.167391
437. "embedding.407_of_512" 0.167391
438. "embedding.017_of_512" 0.167391
439. "embedding.047_of_512" 0.167391
440. "embedding.048_of_512" 0.167391
441. "embedding.128_of_512" 0.167391
442. "embedding.129_of_512" 0.167391
443. "embedding.148_of_512" 0.167391
444. "embedding.248_of_512" 0.167391
445. "embedding.367_of_512" 0.167391
446. "embedding.374_of_512" 0.167391
447. "embedding.398_of_512" 0.167391
448. "embedding.419_of_512" 0.167391
449. "embedding.490_of_512" 0.167391
450. "embedding.506_of_512" 0.167391
451. "embedding.509_of_512" 0.167391
452. "embedding.019_of_512" 0.167390
453. "embedding.103_of_512" 0.167390
454. "embedding.108_of_512" 0.167390
455. "embedding.117_of_512" 0.167390
456. "embedding.121_of_512" 0.167390
457. "embedding.127_of_512" 0.167390
458. "embedding.166_of_512" 0.167390
459. "embedding.181_of_512" 0.167390
460. "embedding.203_of_512" 0.167390
461. "embedding.270_of_512" 0.167390
462. "embedding.273_of_512" 0.167390
463. "embedding.275_of_512" 0.167390
464. "embedding.299_of_512" 0.167390
465. "embedding.318_of_512" 0.167390
466. "embedding.377_of_512" 0.167390
467. "embedding.392_of_512" 0.167390
468. "embedding.435_of_512" 0.167390
469. "embedding.453_of_512" 0.167390
470. "embedding.491_of_512" 0.167390
471. "embedding.162_of_512" 0.167390
472. "embedding.285_of_512" 0.167390
473. "embedding.295_of_512" 0.167390
474. "embedding.395_of_512" 0.167390
475. "embedding.446_of_512" 0.167390
476. "embedding.486_of_512" 0.167390
1. "embedding.171_of_512" 18.000000 ################
2. "embedding.111_of_512" 13.000000 ###########
3. "embedding.036_of_512" 4.000000 ##
4. "embedding.408_of_512" 4.000000 ##
5. "embedding.058_of_512" 3.000000 #
6. "embedding.167_of_512" 3.000000 #
7. "embedding.276_of_512" 3.000000 #
8. "embedding.332_of_512" 3.000000 #
9. "embedding.333_of_512" 3.000000 #
10. "embedding.041_of_512" 2.000000
11. "embedding.043_of_512" 2.000000
12. "embedding.063_of_512" 2.000000
13. "embedding.118_of_512" 2.000000
14. "embedding.237_of_512" 2.000000
15. "embedding.302_of_512" 2.000000
16. "embedding.307_of_512" 2.000000
17. "embedding.329_of_512" 2.000000
18. "embedding.343_of_512" 2.000000
19. "embedding.346_of_512" 2.000000
20. "embedding.350_of_512" 2.000000
21. "embedding.365_of_512" 2.000000
22. "embedding.384_of_512" 2.000000
23. "embedding.044_of_512" 1.000000
24. "embedding.046_of_512" 1.000000
25. "embedding.052_of_512" 1.000000
26. "embedding.079_of_512" 1.000000
27. "embedding.132_of_512" 1.000000
28. "embedding.140_of_512" 1.000000
29. "embedding.175_of_512" 1.000000
30. "embedding.182_of_512" 1.000000
31. "embedding.184_of_512" 1.000000
32. "embedding.188_of_512" 1.000000
33. "embedding.227_of_512" 1.000000
34. "embedding.287_of_512" 1.000000
35. "embedding.291_of_512" 1.000000
36. "embedding.297_of_512" 1.000000
37. "embedding.306_of_512" 1.000000
38. "embedding.319_of_512" 1.000000
39. "embedding.342_of_512" 1.000000
40. "embedding.352_of_512" 1.000000
41. "embedding.363_of_512" 1.000000
42. "embedding.511_of_512" 1.000000
1. "embedding.111_of_512" 103.000000 ################
2. "embedding.171_of_512" 90.000000 #############
3. "embedding.022_of_512" 50.000000 #######
4. "embedding.291_of_512" 47.000000 #######
5. "embedding.384_of_512" 44.000000 ######
6. "embedding.188_of_512" 38.000000 #####
7. "embedding.365_of_512" 35.000000 #####
8. "embedding.092_of_512" 34.000000 #####
9. "embedding.132_of_512" 33.000000 #####
10. "embedding.276_of_512" 32.000000 ####
11. "embedding.140_of_512" 31.000000 ####
12. "embedding.408_of_512" 31.000000 ####
13. "embedding.065_of_512" 29.000000 ####
14. "embedding.184_of_512" 29.000000 ####
15. "embedding.350_of_512" 28.000000 ####
16. "embedding.307_of_512" 26.000000 ###
17. "embedding.063_of_512" 24.000000 ###
18. "embedding.182_of_512" 24.000000 ###
19. "embedding.343_of_512" 24.000000 ###
20. "embedding.292_of_512" 22.000000 ###
21. "embedding.289_of_512" 21.000000 ###
22. "embedding.297_of_512" 20.000000 ##
23. "embedding.194_of_512" 18.000000 ##
24. "embedding.376_of_512" 18.000000 ##
25. "embedding.503_of_512" 18.000000 ##
26. "embedding.041_of_512" 17.000000 ##
27. "embedding.195_of_512" 17.000000 ##
28. "embedding.332_of_512" 17.000000 ##
29. "embedding.333_of_512" 17.000000 ##
30. "embedding.342_of_512" 17.000000 ##
31. "embedding.368_of_512" 17.000000 ##
32. "embedding.511_of_512" 17.000000 ##
33. "embedding.227_of_512" 16.000000 ##
34. "embedding.363_of_512" 16.000000 ##
35. "embedding.478_of_512" 16.000000 ##
36. "embedding.494_of_512" 16.000000 ##
37. "embedding.058_of_512" 15.000000 ##
38. "embedding.118_of_512" 15.000000 ##
39. "embedding.240_of_512" 15.000000 ##
40. "embedding.130_of_512" 14.000000 ##
41. "embedding.152_of_512" 14.000000 ##
42. "embedding.169_of_512" 14.000000 ##
43. "embedding.173_of_512" 14.000000 ##
44. "embedding.319_of_512" 14.000000 ##
45. "embedding.356_of_512" 14.000000 ##
46. "embedding.427_of_512" 14.000000 ##
47. "embedding.471_of_512" 14.000000 ##
48. "embedding.066_of_512" 13.000000 #
49. "embedding.219_of_512" 13.000000 #
50. "embedding.244_of_512" 13.000000 #
51. "embedding.281_of_512" 13.000000 #
52. "embedding.283_of_512" 13.000000 #
53. "embedding.329_of_512" 13.000000 #
54. "embedding.424_of_512" 13.000000 #
55. "embedding.449_of_512" 13.000000 #
56. "embedding.071_of_512" 12.000000 #
57. "embedding.175_of_512" 12.000000 #
58. "embedding.185_of_512" 12.000000 #
59. "embedding.205_of_512" 12.000000 #
60. "embedding.243_of_512" 12.000000 #
61. "embedding.296_of_512" 12.000000 #
62. "embedding.308_of_512" 12.000000 #
63. "embedding.346_of_512" 12.000000 #
64. "embedding.012_of_512" 11.000000 #
65. "embedding.020_of_512" 11.000000 #
66. "embedding.050_of_512" 11.000000 #
67. "embedding.126_of_512" 11.000000 #
68. "embedding.204_of_512" 11.000000 #
69. "embedding.302_of_512" 11.000000 #
70. "embedding.399_of_512" 11.000000 #
71. "embedding.483_of_512" 11.000000 #
72. "embedding.498_of_512" 11.000000 #
73. "embedding.502_of_512" 11.000000 #
74. "embedding.036_of_512" 10.000000 #
75. "embedding.079_of_512" 10.000000 #
76. "embedding.116_of_512" 10.000000 #
77. "embedding.215_of_512" 10.000000 #
78. "embedding.251_of_512" 10.000000 #
79. "embedding.265_of_512" 10.000000 #
80. "embedding.358_of_512" 10.000000 #
81. "embedding.375_of_512" 10.000000 #
82. "embedding.437_of_512" 10.000000 #
83. "embedding.444_of_512" 10.000000 #
84. "embedding.030_of_512" 9.000000 #
85. "embedding.068_of_512" 9.000000 #
86. "embedding.076_of_512" 9.000000 #
87. "embedding.093_of_512" 9.000000 #
88. "embedding.142_of_512" 9.000000 #
89. "embedding.146_of_512" 9.000000 #
90. "embedding.167_of_512" 9.000000 #
91. "embedding.306_of_512" 9.000000 #
92. "embedding.327_of_512" 9.000000 #
93. "embedding.416_of_512" 9.000000 #
94. "embedding.433_of_512" 9.000000 #
95. "embedding.505_of_512" 9.000000 #
96. "embedding.044_of_512" 8.000000 #
97. "embedding.054_of_512" 8.000000 #
98. "embedding.090_of_512" 8.000000 #
99. "embedding.098_of_512" 8.000000 #
100. "embedding.191_of_512" 8.000000 #
101. "embedding.217_of_512" 8.000000 #
102. "embedding.233_of_512" 8.000000 #
103. "embedding.268_of_512" 8.000000 #
104. "embedding.362_of_512" 8.000000 #
105. "embedding.378_of_512" 8.000000 #
106. "embedding.443_of_512" 8.000000 #
107. "embedding.469_of_512" 8.000000 #
108. "embedding.043_of_512" 7.000000
109. "embedding.052_of_512" 7.000000
110. "embedding.061_of_512" 7.000000
111. "embedding.067_of_512" 7.000000
112. "embedding.069_of_512" 7.000000
113. "embedding.082_of_512" 7.000000
114. "embedding.113_of_512" 7.000000
115. "embedding.114_of_512" 7.000000
116. "embedding.122_of_512" 7.000000
117. "embedding.158_of_512" 7.000000
118. "embedding.190_of_512" 7.000000
119. "embedding.202_of_512" 7.000000
120. "embedding.290_of_512" 7.000000
121. "embedding.325_of_512" 7.000000
122. "embedding.402_of_512" 7.000000
123. "embedding.423_of_512" 7.000000
124. "embedding.463_of_512" 7.000000
125. "embedding.018_of_512" 6.000000
126. "embedding.040_of_512" 6.000000
127. "embedding.042_of_512" 6.000000
128. "embedding.060_of_512" 6.000000
129. "embedding.062_of_512" 6.000000
130. "embedding.064_of_512" 6.000000
131. "embedding.086_of_512" 6.000000
132. "embedding.119_of_512" 6.000000
133. "embedding.135_of_512" 6.000000
134. "embedding.141_of_512" 6.000000
135. "embedding.165_of_512" 6.000000
136. "embedding.177_of_512" 6.000000
137. "embedding.192_of_512" 6.000000
138. "embedding.197_of_512" 6.000000
139. "embedding.225_of_512" 6.000000
140. "embedding.232_of_512" 6.000000
141. "embedding.236_of_512" 6.000000
142. "embedding.237_of_512" 6.000000
143. "embedding.255_of_512" 6.000000
144. "embedding.256_of_512" 6.000000
145. "embedding.267_of_512" 6.000000
146. "embedding.279_of_512" 6.000000
147. "embedding.284_of_512" 6.000000
148. "embedding.370_of_512" 6.000000
149. "embedding.372_of_512" 6.000000
150. "embedding.382_of_512" 6.000000
151. "embedding.405_of_512" 6.000000
152. "embedding.417_of_512" 6.000000
153. "embedding.418_of_512" 6.000000
154. "embedding.462_of_512" 6.000000
155. "embedding.470_of_512" 6.000000
156. "embedding.480_of_512" 6.000000
157. "embedding.482_of_512" 6.000000
158. "embedding.499_of_512" 6.000000
159. "embedding.011_of_512" 5.000000
160. "embedding.025_of_512" 5.000000
161. "embedding.039_of_512" 5.000000
162. "embedding.051_of_512" 5.000000
163. "embedding.057_of_512" 5.000000
164. "embedding.072_of_512" 5.000000
165. "embedding.080_of_512" 5.000000
166. "embedding.081_of_512" 5.000000
167. "embedding.095_of_512" 5.000000
168. "embedding.101_of_512" 5.000000
169. "embedding.133_of_512" 5.000000
170. "embedding.134_of_512" 5.000000
171. "embedding.136_of_512" 5.000000
172. "embedding.155_of_512" 5.000000
173. "embedding.180_of_512" 5.000000
174. "embedding.187_of_512" 5.000000
175. "embedding.193_of_512" 5.000000
176. "embedding.200_of_512" 5.000000
177. "embedding.209_of_512" 5.000000
178. "embedding.212_of_512" 5.000000
179. "embedding.213_of_512" 5.000000
180. "embedding.216_of_512" 5.000000
181. "embedding.221_of_512" 5.000000
182. "embedding.235_of_512" 5.000000
183. "embedding.245_of_512" 5.000000
184. "embedding.263_of_512" 5.000000
185. "embedding.288_of_512" 5.000000
186. "embedding.305_of_512" 5.000000
187. "embedding.314_of_512" 5.000000
188. "embedding.354_of_512" 5.000000
189. "embedding.359_of_512" 5.000000
190. "embedding.366_of_512" 5.000000
191. "embedding.381_of_512" 5.000000
192. "embedding.411_of_512" 5.000000
193. "embedding.412_of_512" 5.000000
194. "embedding.415_of_512" 5.000000
195. "embedding.421_of_512" 5.000000
196. "embedding.454_of_512" 5.000000
197. "embedding.465_of_512" 5.000000
198. "embedding.474_of_512" 5.000000
199. "embedding.484_of_512" 5.000000
200. "embedding.492_of_512" 5.000000
201. "embedding.000_of_512" 4.000000
202. "embedding.024_of_512" 4.000000
203. "embedding.027_of_512" 4.000000
204. "embedding.046_of_512" 4.000000
205. "embedding.059_of_512" 4.000000
206. "embedding.074_of_512" 4.000000
207. "embedding.075_of_512" 4.000000
208. "embedding.078_of_512" 4.000000
209. "embedding.084_of_512" 4.000000
210. "embedding.089_of_512" 4.000000
211. "embedding.107_of_512" 4.000000
212. "embedding.109_of_512" 4.000000
213. "embedding.123_of_512" 4.000000
214. "embedding.143_of_512" 4.000000
215. "embedding.150_of_512" 4.000000
216. "embedding.153_of_512" 4.000000
217. "embedding.159_of_512" 4.000000
218. "embedding.160_of_512" 4.000000
219. "embedding.161_of_512" 4.000000
220. "embedding.189_of_512" 4.000000
221. "embedding.210_of_512" 4.000000
222. "embedding.224_of_512" 4.000000
223. "embedding.247_of_512" 4.000000
224. "embedding.250_of_512" 4.000000
225. "embedding.261_of_512" 4.000000
226. "embedding.262_of_512" 4.000000
227. "embedding.264_of_512" 4.000000
228. "embedding.293_of_512" 4.000000
229. "embedding.294_of_512" 4.000000
230. "embedding.303_of_512" 4.000000
231. "embedding.317_of_512" 4.000000
232. "embedding.320_of_512" 4.000000
233. "embedding.322_of_512" 4.000000
234. "embedding.323_of_512" 4.000000
235. "embedding.324_of_512" 4.000000
236. "embedding.338_of_512" 4.000000
237. "embedding.345_of_512" 4.000000
238. "embedding.371_of_512" 4.000000
239. "embedding.385_of_512" 4.000000
240. "embedding.391_of_512" 4.000000
241. "embedding.410_of_512" 4.000000
242. "embedding.442_of_512" 4.000000
243. "embedding.451_of_512" 4.000000
244. "embedding.458_of_512" 4.000000
245. "embedding.475_of_512" 4.000000
246. "embedding.477_of_512" 4.000000
247. "embedding.479_of_512" 4.000000
248. "embedding.493_of_512" 4.000000
249. "embedding.496_of_512" 4.000000
250. "embedding.508_of_512" 4.000000
251. "embedding.001_of_512" 3.000000
252. "embedding.004_of_512" 3.000000
253. "embedding.005_of_512" 3.000000
254. "embedding.006_of_512" 3.000000
255. "embedding.010_of_512" 3.000000
256. "embedding.014_of_512" 3.000000
257. "embedding.026_of_512" 3.000000
258. "embedding.038_of_512" 3.000000
259. "embedding.053_of_512" 3.000000
260. "embedding.070_of_512" 3.000000
261. "embedding.085_of_512" 3.000000
262. "embedding.087_of_512" 3.000000
263. "embedding.096_of_512" 3.000000
264. "embedding.097_of_512" 3.000000
265. "embedding.100_of_512" 3.000000
266. "embedding.104_of_512" 3.000000
267. "embedding.105_of_512" 3.000000
268. "embedding.110_of_512" 3.000000
269. "embedding.125_of_512" 3.000000
270. "embedding.131_of_512" 3.000000
271. "embedding.137_of_512" 3.000000
272. "embedding.138_of_512" 3.000000
273. "embedding.144_of_512" 3.000000
274. "embedding.151_of_512" 3.000000
275. "embedding.163_of_512" 3.000000
276. "embedding.164_of_512" 3.000000
277. "embedding.168_of_512" 3.000000
278. "embedding.174_of_512" 3.000000
279. "embedding.179_of_512" 3.000000
280. "embedding.206_of_512" 3.000000
281. "embedding.208_of_512" 3.000000
282. "embedding.220_of_512" 3.000000
283. "embedding.223_of_512" 3.000000
284. "embedding.228_of_512" 3.000000
285. "embedding.230_of_512" 3.000000
286. "embedding.238_of_512" 3.000000
287. "embedding.239_of_512" 3.000000
288. "embedding.260_of_512" 3.000000
289. "embedding.266_of_512" 3.000000
290. "embedding.269_of_512" 3.000000
291. "embedding.280_of_512" 3.000000
292. "embedding.286_of_512" 3.000000
293. "embedding.287_of_512" 3.000000
294. "embedding.309_of_512" 3.000000
295. "embedding.312_of_512" 3.000000
296. "embedding.315_of_512" 3.000000
297. "embedding.326_of_512" 3.000000
298. "embedding.336_of_512" 3.000000
299. "embedding.337_of_512" 3.000000
300. "embedding.340_of_512" 3.000000
301. "embedding.341_of_512" 3.000000
302. "embedding.348_of_512" 3.000000
303. "embedding.351_of_512" 3.000000
304. "embedding.352_of_512" 3.000000
305. "embedding.361_of_512" 3.000000
306. "embedding.388_of_512" 3.000000
307. "embedding.389_of_512" 3.000000
308. "embedding.390_of_512" 3.000000
309. "embedding.400_of_512" 3.000000
310. "embedding.403_of_512" 3.000000
311. "embedding.406_of_512" 3.000000
312. "embedding.422_of_512" 3.000000
313. "embedding.428_of_512" 3.000000
314. "embedding.431_of_512" 3.000000
315. "embedding.438_of_512" 3.000000
316. "embedding.440_of_512" 3.000000
317. "embedding.452_of_512" 3.000000
318. "embedding.456_of_512" 3.000000
319. "embedding.461_of_512" 3.000000
320. "embedding.466_of_512" 3.000000
321. "embedding.472_of_512" 3.000000
322. "embedding.476_of_512" 3.000000
323. "embedding.487_of_512" 3.000000
324. "embedding.510_of_512" 3.000000
325. "embedding.002_of_512" 2.000000
326. "embedding.003_of_512" 2.000000
327. "embedding.007_of_512" 2.000000
328. "embedding.008_of_512" 2.000000
329. "embedding.009_of_512" 2.000000
330. "embedding.016_of_512" 2.000000
331. "embedding.021_of_512" 2.000000
332. "embedding.023_of_512" 2.000000
333. "embedding.028_of_512" 2.000000
334. "embedding.032_of_512" 2.000000
335. "embedding.033_of_512" 2.000000
336. "embedding.035_of_512" 2.000000
337. "embedding.037_of_512" 2.000000
338. "embedding.045_of_512" 2.000000
339. "embedding.049_of_512" 2.000000
340. "embedding.055_of_512" 2.000000
341. "embedding.056_of_512" 2.000000
342. "embedding.073_of_512" 2.000000
343. "embedding.077_of_512" 2.000000
344. "embedding.083_of_512" 2.000000
345. "embedding.091_of_512" 2.000000
346. "embedding.094_of_512" 2.000000
347. "embedding.102_of_512" 2.000000
348. "embedding.106_of_512" 2.000000
349. "embedding.115_of_512" 2.000000
350. "embedding.120_of_512" 2.000000
351. "embedding.124_of_512" 2.000000
352. "embedding.139_of_512" 2.000000
353. "embedding.145_of_512" 2.000000
354. "embedding.156_of_512" 2.000000
355. "embedding.157_of_512" 2.000000
356. "embedding.170_of_512" 2.000000
357. "embedding.172_of_512" 2.000000
358. "embedding.183_of_512" 2.000000
359. "embedding.198_of_512" 2.000000
360. "embedding.199_of_512" 2.000000
361. "embedding.207_of_512" 2.000000
362. "embedding.211_of_512" 2.000000
363. "embedding.218_of_512" 2.000000
364. "embedding.226_of_512" 2.000000
365. "embedding.229_of_512" 2.000000
366. "embedding.231_of_512" 2.000000
367. "embedding.234_of_512" 2.000000
368. "embedding.257_of_512" 2.000000
369. "embedding.258_of_512" 2.000000
370. "embedding.259_of_512" 2.000000
371. "embedding.271_of_512" 2.000000
372. "embedding.272_of_512" 2.000000
373. "embedding.274_of_512" 2.000000
374. "embedding.277_of_512" 2.000000
375. "embedding.278_of_512" 2.000000
376. "embedding.301_of_512" 2.000000
377. "embedding.304_of_512" 2.000000
378. "embedding.313_of_512" 2.000000
379. "embedding.316_of_512" 2.000000
380. "embedding.321_of_512" 2.000000
381. "embedding.328_of_512" 2.000000
382. "embedding.330_of_512" 2.000000
383. "embedding.331_of_512" 2.000000
384. "embedding.344_of_512" 2.000000
385. "embedding.347_of_512" 2.000000
386. "embedding.353_of_512" 2.000000
387. "embedding.357_of_512" 2.000000
388. "embedding.360_of_512" 2.000000
389. "embedding.364_of_512" 2.000000
390. "embedding.373_of_512" 2.000000
391. "embedding.379_of_512" 2.000000
392. "embedding.380_of_512" 2.000000
393. "embedding.386_of_512" 2.000000
394. "embedding.404_of_512" 2.000000
395. "embedding.425_of_512" 2.000000
396. "embedding.426_of_512" 2.000000
397. "embedding.430_of_512" 2.000000
398. "embedding.432_of_512" 2.000000
399. "embedding.434_of_512" 2.000000
400. "embedding.439_of_512" 2.000000
401. "embedding.441_of_512" 2.000000
402. "embedding.445_of_512" 2.000000
403. "embedding.447_of_512" 2.000000
404. "embedding.457_of_512" 2.000000
405. "embedding.459_of_512" 2.000000
406. "embedding.460_of_512" 2.000000
407. "embedding.464_of_512" 2.000000
408. "embedding.467_of_512" 2.000000
409. "embedding.468_of_512" 2.000000
410. "embedding.489_of_512" 2.000000
411. "embedding.500_of_512" 2.000000
412. "embedding.501_of_512" 2.000000
413. "embedding.507_of_512" 2.000000
414. "embedding.015_of_512" 1.000000
415. "embedding.017_of_512" 1.000000
416. "embedding.019_of_512" 1.000000
417. "embedding.029_of_512" 1.000000
418. "embedding.031_of_512" 1.000000
419. "embedding.034_of_512" 1.000000
420. "embedding.047_of_512" 1.000000
421. "embedding.048_of_512" 1.000000
422. "embedding.103_of_512" 1.000000
423. "embedding.108_of_512" 1.000000
424. "embedding.117_of_512" 1.000000
425. "embedding.121_of_512" 1.000000
426. "embedding.127_of_512" 1.000000
427. "embedding.128_of_512" 1.000000
428. "embedding.129_of_512" 1.000000
429. "embedding.148_of_512" 1.000000
430. "embedding.149_of_512" 1.000000
431. "embedding.154_of_512" 1.000000
432. "embedding.162_of_512" 1.000000
433. "embedding.166_of_512" 1.000000
434. "embedding.178_of_512" 1.000000
435. "embedding.181_of_512" 1.000000
436. "embedding.196_of_512" 1.000000
437. "embedding.203_of_512" 1.000000
438. "embedding.214_of_512" 1.000000
439. "embedding.248_of_512" 1.000000
440. "embedding.252_of_512" 1.000000
441. "embedding.270_of_512" 1.000000
442. "embedding.273_of_512" 1.000000
443. "embedding.275_of_512" 1.000000
444. "embedding.282_of_512" 1.000000
445. "embedding.285_of_512" 1.000000
446. "embedding.295_of_512" 1.000000
447. "embedding.299_of_512" 1.000000
448. "embedding.300_of_512" 1.000000
449. "embedding.310_of_512" 1.000000
450. "embedding.318_of_512" 1.000000
451. "embedding.335_of_512" 1.000000
452. "embedding.367_of_512" 1.000000
453. "embedding.369_of_512" 1.000000
454. "embedding.374_of_512" 1.000000
455. "embedding.377_of_512" 1.000000
456. "embedding.383_of_512" 1.000000
457. "embedding.392_of_512" 1.000000
458. "embedding.394_of_512" 1.000000
459. "embedding.395_of_512" 1.000000
460. "embedding.398_of_512" 1.000000
461. "embedding.401_of_512" 1.000000
462. "embedding.407_of_512" 1.000000
463. "embedding.409_of_512" 1.000000
464. "embedding.413_of_512" 1.000000
465. "embedding.419_of_512" 1.000000
466. "embedding.435_of_512" 1.000000
467. "embedding.446_of_512" 1.000000
468. "embedding.453_of_512" 1.000000
469. "embedding.481_of_512" 1.000000
470. "embedding.485_of_512" 1.000000
471. "embedding.486_of_512" 1.000000
472. "embedding.490_of_512" 1.000000
473. "embedding.491_of_512" 1.000000
474. "embedding.495_of_512" 1.000000
475. "embedding.506_of_512" 1.000000
476. "embedding.509_of_512" 1.000000
1. "embedding.171_of_512" 17163.261611 ################
2. "embedding.111_of_512" 9222.136584 ########
3. "embedding.291_of_512" 2011.546091 #
4. "embedding.384_of_512" 1440.158334 #
5. "embedding.022_of_512" 1024.883766
6. "embedding.365_of_512" 798.629422
7. "embedding.188_of_512" 593.985603
8. "embedding.092_of_512" 439.705607
9. "embedding.132_of_512" 434.868580
10. "embedding.276_of_512" 420.460029
11. "embedding.408_of_512" 382.588801
12. "embedding.140_of_512" 377.887316
13. "embedding.182_of_512" 334.849162
14. "embedding.307_of_512" 326.094298
15. "embedding.289_of_512" 307.450534
16. "embedding.065_of_512" 305.324133
17. "embedding.184_of_512" 296.966936
18. "embedding.350_of_512" 247.403096
19. "embedding.478_of_512" 224.317990
20. "embedding.343_of_512" 216.826511
21. "embedding.368_of_512" 212.126720
22. "embedding.063_of_512" 196.805053
23. "embedding.342_of_512" 183.624144
24. "embedding.427_of_512" 178.465117
25. "embedding.333_of_512" 174.503619
26. "embedding.297_of_512" 164.345234
27. "embedding.503_of_512" 164.037252
28. "embedding.152_of_512" 153.431485
29. "embedding.494_of_512" 144.618493
30. "embedding.443_of_512" 140.130437
31. "embedding.142_of_512" 138.160709
32. "embedding.346_of_512" 136.212783
33. "embedding.195_of_512" 134.345982
34. "embedding.194_of_512" 134.128821
35. "embedding.424_of_512" 130.271760
36. "embedding.356_of_512" 129.947941
37. "embedding.283_of_512" 129.362391
38. "embedding.292_of_512" 124.217120
39. "embedding.041_of_512" 123.223807
40. "embedding.219_of_512" 122.543446
41. "embedding.483_of_512" 117.867249
42. "embedding.281_of_512" 114.859912
43. "embedding.376_of_512" 109.000620
44. "embedding.363_of_512" 108.860656
45. "embedding.240_of_512" 106.875148
46. "embedding.066_of_512" 105.042712
47. "embedding.169_of_512" 105.037665
48. "embedding.375_of_512" 103.174323
49. "embedding.302_of_512" 99.819298
50. "embedding.308_of_512" 96.264454
51. "embedding.319_of_512" 93.653431
52. "embedding.093_of_512" 91.685980
53. "embedding.243_of_512" 91.042571
54. "embedding.130_of_512" 91.009409
55. "embedding.185_of_512" 87.680004
56. "embedding.227_of_512" 87.044095
57. "embedding.204_of_512" 87.034383
58. "embedding.511_of_512" 84.744629
59. "embedding.399_of_512" 82.595387
60. "embedding.118_of_512" 80.987761
61. "embedding.502_of_512" 78.536618
62. "embedding.449_of_512" 76.502385
63. "embedding.332_of_512" 75.353250
64. "embedding.471_of_512" 75.252508
65. "embedding.071_of_512" 72.293488
66. "embedding.058_of_512" 72.255089
67. "embedding.173_of_512" 69.160309
68. "embedding.052_of_512" 69.043943
69. "embedding.054_of_512" 68.104798
70. "embedding.020_of_512" 67.354732
71. "embedding.329_of_512" 66.390571
72. "embedding.244_of_512" 63.266665
73. "embedding.068_of_512" 62.810135
74. "embedding.251_of_512" 60.992789
75. "embedding.126_of_512" 60.487690
76. "embedding.498_of_512" 58.219495
77. "embedding.036_of_512" 57.881889
78. "embedding.050_of_512" 56.937522
79. "embedding.205_of_512" 56.100721
80. "embedding.296_of_512" 55.948082
81. "embedding.378_of_512" 55.680739
82. "embedding.444_of_512" 54.850502
83. "embedding.358_of_512" 54.629345
84. "embedding.167_of_512" 54.071826
85. "embedding.012_of_512" 53.147135
86. "embedding.069_of_512" 53.089987
87. "embedding.175_of_512" 52.331643
88. "embedding.043_of_512" 51.962867
89. "embedding.217_of_512" 51.956193
90. "embedding.177_of_512" 50.010615
91. "embedding.079_of_512" 49.414317
92. "embedding.437_of_512" 49.269297
93. "embedding.215_of_512" 48.609580
94. "embedding.327_of_512" 47.985854
95. "embedding.463_of_512" 47.883520
96. "embedding.192_of_512" 47.093082
97. "embedding.116_of_512" 46.948855
98. "embedding.146_of_512" 46.826456
99. "embedding.044_of_512" 46.415573
100. "embedding.143_of_512" 45.103554
101. "embedding.076_of_512" 44.452931
102. "embedding.469_of_512" 43.807500
103. "embedding.225_of_512" 43.329500
104. "embedding.233_of_512" 42.304960
105. "embedding.245_of_512" 40.998334
106. "embedding.268_of_512" 40.928178
107. "embedding.314_of_512" 40.694869
108. "embedding.255_of_512" 38.955894
109. "embedding.098_of_512" 38.746280
110. "embedding.505_of_512" 38.626907
111. "embedding.030_of_512" 38.509823
112. "embedding.114_of_512" 37.668527
113. "embedding.306_of_512" 37.350130
114. "embedding.042_of_512" 37.127850
115. "embedding.265_of_512" 37.088925
116. "embedding.433_of_512" 36.785369
117. "embedding.193_of_512" 36.362381
118. "embedding.067_of_512" 35.623289
119. "embedding.150_of_512" 35.106659
120. "embedding.402_of_512" 34.397375
121. "embedding.465_of_512" 34.070292
122. "embedding.113_of_512" 34.020396
123. "embedding.325_of_512" 34.004460
124. "embedding.190_of_512" 33.812829
125. "embedding.247_of_512" 33.571791
126. "embedding.382_of_512" 32.936728
127. "embedding.416_of_512" 32.743171
128. "embedding.072_of_512" 32.739860
129. "embedding.155_of_512" 32.528433
130. "embedding.279_of_512" 32.287712
131. "embedding.417_of_512" 32.038925
132. "embedding.061_of_512" 31.225254
133. "embedding.267_of_512" 30.948638
134. "embedding.362_of_512" 30.790368
135. "embedding.086_of_512" 30.737318
136. "embedding.191_of_512" 30.681134
137. "embedding.366_of_512" 30.374531
138. "embedding.090_of_512" 29.621714
139. "embedding.232_of_512" 28.934306
140. "embedding.082_of_512" 28.929476
141. "embedding.101_of_512" 28.822461
142. "embedding.410_of_512" 28.390060
143. "embedding.026_of_512" 27.886741
144. "embedding.423_of_512" 27.524733
145. "embedding.080_of_512" 27.194503
146. "embedding.180_of_512" 26.866280
147. "embedding.216_of_512" 26.833603
148. "embedding.011_of_512" 26.743176
149. "embedding.236_of_512" 26.659612
150. "embedding.158_of_512" 26.437004
151. "embedding.197_of_512" 26.104442
152. "embedding.480_of_512" 25.937455
153. "embedding.135_of_512" 25.826829
154. "embedding.202_of_512" 25.767675
155. "embedding.256_of_512" 25.752772
156. "embedding.404_of_512" 25.730176
157. "embedding.057_of_512" 25.713334
158. "embedding.263_of_512" 25.631016
159. "embedding.482_of_512" 25.280783
160. "embedding.062_of_512" 25.175564
161. "embedding.284_of_512" 25.075373
162. "embedding.470_of_512" 25.049711
163. "embedding.136_of_512" 25.014152
164. "embedding.040_of_512" 24.838080
165. "embedding.418_of_512" 24.201707
166. "embedding.324_of_512" 23.976690
167. "embedding.109_of_512" 23.801806
168. "embedding.391_of_512" 23.550975
169. "embedding.290_of_512" 23.394580
170. "embedding.187_of_512" 23.258433
171. "embedding.405_of_512" 23.122473
172. "embedding.221_of_512" 23.117095
173. "embedding.119_of_512" 22.902634
174. "embedding.025_of_512" 22.826385
175. "embedding.345_of_512" 22.597737
176. "embedding.122_of_512" 22.411945
177. "embedding.039_of_512" 22.320384
178. "embedding.060_of_512" 22.304902
179. "embedding.262_of_512" 22.021096
180. "embedding.381_of_512" 21.759861
181. "embedding.165_of_512" 21.728633
182. "embedding.499_of_512" 21.633024
183. "embedding.074_of_512" 21.594031
184. "embedding.051_of_512" 21.301531
185. "embedding.237_of_512" 21.156419
186. "embedding.371_of_512" 21.131870
187. "embedding.508_of_512" 21.000655
188. "embedding.317_of_512" 20.679063
189. "embedding.421_of_512" 20.666468
190. "embedding.235_of_512" 20.389141
191. "embedding.239_of_512" 19.915094
192. "embedding.027_of_512" 19.772421
193. "embedding.095_of_512" 19.266230
194. "embedding.492_of_512" 18.949160
195. "embedding.070_of_512" 18.817319
196. "embedding.370_of_512" 18.771609
197. "embedding.354_of_512" 18.714025
198. "embedding.089_of_512" 18.706402
199. "embedding.064_of_512" 18.472540
200. "embedding.006_of_512" 18.288014
201. "embedding.212_of_512" 18.248336
202. "embedding.018_of_512" 18.210169
203. "embedding.372_of_512" 18.089667
204. "embedding.046_of_512" 17.988413
205. "embedding.200_of_512" 17.920917
206. "embedding.141_of_512" 17.893822
207. "embedding.462_of_512" 17.821465
208. "embedding.081_of_512" 17.694298
209. "embedding.107_of_512" 17.566795
210. "embedding.305_of_512" 17.504576
211. "embedding.261_of_512" 17.480185
212. "embedding.087_of_512" 17.274760
213. "embedding.220_of_512" 17.223686
214. "embedding.000_of_512" 17.153120
215. "embedding.496_of_512" 17.089169
216. "embedding.210_of_512" 17.079963
217. "embedding.320_of_512" 17.063082
218. "embedding.134_of_512" 16.808995
219. "embedding.359_of_512" 16.806446
220. "embedding.293_of_512" 16.792584
221. "embedding.133_of_512" 16.714364
222. "embedding.138_of_512" 16.382440
223. "embedding.452_of_512" 16.213482
224. "embedding.075_of_512" 15.913921
225. "embedding.059_of_512" 15.839770
226. "embedding.415_of_512" 15.773601
227. "embedding.174_of_512" 15.639605
228. "embedding.024_of_512" 15.292290
229. "embedding.105_of_512" 15.232853
230. "embedding.287_of_512" 15.135759
231. "embedding.336_of_512" 15.119497
232. "embedding.510_of_512" 15.114059
233. "embedding.161_of_512" 14.983866
234. "embedding.209_of_512" 14.911226
235. "embedding.078_of_512" 14.860597
236. "embedding.123_of_512" 14.738864
237. "embedding.228_of_512" 14.704528
238. "embedding.280_of_512" 14.700699
239. "embedding.250_of_512" 14.622368
240. "embedding.412_of_512" 14.549791
241. "embedding.084_of_512" 14.457128
242. "embedding.224_of_512" 14.325537
243. "embedding.323_of_512" 14.301669
244. "embedding.458_of_512" 14.175333
245. "embedding.385_of_512" 14.152160
246. "embedding.411_of_512" 14.127793
247. "embedding.361_of_512" 14.030305
248. "embedding.484_of_512" 13.869317
249. "embedding.257_of_512" 13.863068
250. "embedding.179_of_512" 13.750212
251. "embedding.442_of_512" 13.720752
252. "embedding.223_of_512" 13.591060
253. "embedding.451_of_512" 13.567758
254. "embedding.163_of_512" 13.495614
255. "embedding.341_of_512" 13.326650
256. "embedding.474_of_512" 13.315612
257. "embedding.005_of_512" 13.226211
258. "embedding.286_of_512" 13.069766
259. "embedding.264_of_512" 13.052572
260. "embedding.479_of_512" 12.980791
261. "embedding.014_of_512" 12.954876
262. "embedding.110_of_512" 12.938494
263. "embedding.131_of_512" 12.828036
264. "embedding.489_of_512" 12.754188
265. "embedding.461_of_512" 12.693593
266. "embedding.288_of_512" 12.610884
267. "embedding.328_of_512" 12.424237
268. "embedding.213_of_512" 12.373543
269. "embedding.160_of_512" 12.324912
270. "embedding.125_of_512" 12.296855
271. "embedding.476_of_512" 12.159989
272. "embedding.049_of_512" 12.135087
273. "embedding.440_of_512" 12.127148
274. "embedding.096_of_512" 12.009432
275. "embedding.318_of_512" 11.805688
276. "embedding.038_of_512" 11.743252
277. "embedding.390_of_512" 11.634256
278. "embedding.211_of_512" 11.608288
279. "embedding.294_of_512" 11.542138
280. "embedding.348_of_512" 11.415023
281. "embedding.454_of_512" 11.385024
282. "embedding.144_of_512" 11.308329
283. "embedding.431_of_512" 11.241699
284. "embedding.326_of_512" 11.238328
285. "embedding.475_of_512" 11.232105
286. "embedding.400_of_512" 11.220325
287. "embedding.164_of_512" 11.212532
288. "embedding.373_of_512" 11.167605
289. "embedding.303_of_512" 11.137827
290. "embedding.466_of_512" 10.996574
291. "embedding.322_of_512" 10.987503
292. "embedding.104_of_512" 10.878272
293. "embedding.338_of_512" 10.680139
294. "embedding.278_of_512" 10.661013
295. "embedding.439_of_512" 10.625585
296. "embedding.493_of_512" 10.580044
297. "embedding.386_of_512" 10.416147
298. "embedding.001_of_512" 10.330815
299. "embedding.477_of_512" 10.215354
300. "embedding.274_of_512" 10.195050
301. "embedding.189_of_512" 10.034064
302. "embedding.501_of_512" 9.712892
303. "embedding.023_of_512" 9.704768
304. "embedding.337_of_512" 9.698276
305. "embedding.438_of_512" 9.515732
306. "embedding.238_of_512" 9.445766
307. "embedding.425_of_512" 9.411107
308. "embedding.208_of_512" 9.390540
309. "embedding.315_of_512" 9.389285
310. "embedding.010_of_512" 9.344792
311. "embedding.434_of_512" 9.321555
312. "embedding.344_of_512" 9.305979
313. "embedding.151_of_512" 9.257123
314. "embedding.330_of_512" 9.217059
315. "embedding.406_of_512" 9.214288
316. "embedding.340_of_512" 9.188421
317. "embedding.159_of_512" 9.052036
318. "embedding.137_of_512" 9.038237
319. "embedding.056_of_512" 8.905088
320. "embedding.459_of_512" 8.895027
321. "embedding.007_of_512" 8.830927
322. "embedding.353_of_512" 8.829274
323. "embedding.388_of_512" 8.813356
324. "embedding.313_of_512" 8.809157
325. "embedding.456_of_512" 8.672220
326. "embedding.229_of_512" 8.619430
327. "embedding.321_of_512" 8.604794
328. "embedding.266_of_512" 8.576367
329. "embedding.379_of_512" 8.476270
330. "embedding.100_of_512" 8.439626
331. "embedding.053_of_512" 8.413065
332. "embedding.468_of_512" 8.388147
333. "embedding.352_of_512" 8.382706
334. "embedding.009_of_512" 8.268669
335. "embedding.234_of_512" 8.223046
336. "embedding.447_of_512" 8.218239
337. "embedding.269_of_512" 8.212989
338. "embedding.301_of_512" 8.075671
339. "embedding.422_of_512" 8.063095
340. "embedding.156_of_512" 8.042654
341. "embedding.428_of_512" 8.035930
342. "embedding.153_of_512" 8.029318
343. "embedding.003_of_512" 7.987826
344. "embedding.106_of_512" 7.969378
345. "embedding.507_of_512" 7.824125
346. "embedding.312_of_512" 7.728918
347. "embedding.258_of_512" 7.615799
348. "embedding.226_of_512" 7.458633
349. "embedding.430_of_512" 7.401733
350. "embedding.168_of_512" 7.393336
351. "embedding.035_of_512" 7.387460
352. "embedding.487_of_512" 7.330754
353. "embedding.309_of_512" 7.330402
354. "embedding.033_of_512" 7.309564
355. "embedding.304_of_512" 7.307759
356. "embedding.091_of_512" 7.254614
357. "embedding.102_of_512" 7.214308
358. "embedding.357_of_512" 7.210662
359. "embedding.157_of_512" 7.165839
360. "embedding.094_of_512" 7.165105
361. "embedding.472_of_512" 7.144578
362. "embedding.139_of_512" 7.064992
363. "embedding.377_of_512" 7.042943
364. "embedding.380_of_512" 7.041339
365. "embedding.272_of_512" 7.041162
366. "embedding.004_of_512" 7.035992
367. "embedding.085_of_512" 6.984889
368. "embedding.230_of_512" 6.945414
369. "embedding.432_of_512" 6.943797
370. "embedding.389_of_512" 6.881828
371. "embedding.218_of_512" 6.842948
372. "embedding.008_of_512" 6.803792
373. "embedding.097_of_512" 6.738676
374. "embedding.445_of_512" 6.670537
375. "embedding.037_of_512" 6.596426
376. "embedding.002_of_512" 6.565671
377. "embedding.271_of_512" 6.314382
378. "embedding.207_of_512" 6.268816
379. "embedding.103_of_512" 6.197685
380. "embedding.331_of_512" 6.188369
381. "embedding.460_of_512" 6.159943
382. "embedding.347_of_512" 6.139413
383. "embedding.206_of_512" 6.100772
384. "embedding.351_of_512" 5.963090
385. "embedding.441_of_512" 5.958451
386. "embedding.032_of_512" 5.841272
387. "embedding.464_of_512" 5.748941
388. "embedding.172_of_512" 5.668636
389. "embedding.045_of_512" 5.624897
390. "embedding.495_of_512" 5.606474
391. "embedding.231_of_512" 5.488857
392. "embedding.457_of_512" 5.412411
393. "embedding.360_of_512" 5.408576
394. "embedding.124_of_512" 5.398018
395. "embedding.198_of_512" 5.389353
396. "embedding.021_of_512" 5.353755
397. "embedding.277_of_512" 5.305103
398. "embedding.467_of_512" 5.292973
399. "embedding.120_of_512" 5.268761
400. "embedding.506_of_512" 5.188110
401. "embedding.500_of_512" 5.162792
402. "embedding.403_of_512" 5.110633
403. "embedding.083_of_512" 5.093338
404. "embedding.048_of_512" 5.090771
405. "embedding.259_of_512" 4.962671
406. "embedding.491_of_512" 4.955279
407. "embedding.426_of_512" 4.930060
408. "embedding.170_of_512" 4.810002
409. "embedding.162_of_512" 4.763983
410. "embedding.335_of_512" 4.745229
411. "embedding.199_of_512" 4.703493
412. "embedding.260_of_512" 4.625093
413. "embedding.275_of_512" 4.350696
414. "embedding.364_of_512" 4.328313
415. "embedding.183_of_512" 4.213411
416. "embedding.108_of_512" 4.180671
417. "embedding.117_of_512" 4.159432
418. "embedding.300_of_512" 4.080444
419. "embedding.409_of_512" 4.074465
420. "embedding.178_of_512" 4.040530
421. "embedding.310_of_512" 3.940741
422. "embedding.016_of_512" 3.898330
423. "embedding.148_of_512" 3.897144
424. "embedding.419_of_512" 3.813937
425. "embedding.077_of_512" 3.810567
426. "embedding.453_of_512" 3.758238
427. "embedding.019_of_512" 3.755729
428. "embedding.270_of_512" 3.727955
429. "embedding.490_of_512" 3.595743
430. "embedding.398_of_512" 3.510094
431. "embedding.127_of_512" 3.494055
432. "embedding.203_of_512" 3.392086
433. "embedding.446_of_512" 3.358459
434. "embedding.413_of_512" 3.340526
435. "embedding.145_of_512" 3.282636
436. "embedding.128_of_512" 3.217526
437. "embedding.166_of_512" 3.124052
438. "embedding.394_of_512" 3.113460
439. "embedding.316_of_512" 3.064561
440. "embedding.149_of_512" 3.053578
441. "embedding.369_of_512" 2.945368
442. "embedding.028_of_512" 2.810905
443. "embedding.181_of_512" 2.743388
444. "embedding.196_of_512" 2.742419
445. "embedding.407_of_512" 2.738602
446. "embedding.273_of_512" 2.672149
447. "embedding.015_of_512" 2.670484
448. "embedding.295_of_512" 2.441439
449. "embedding.285_of_512" 2.377308
450. "embedding.395_of_512" 2.374672
451. "embedding.485_of_512" 2.341882
452. "embedding.392_of_512" 2.339278
453. "embedding.252_of_512" 2.329295
454. "embedding.154_of_512" 2.317964
455. "embedding.017_of_512" 2.229209
456. "embedding.073_of_512" 2.228032
457. "embedding.129_of_512" 2.217764
458. "embedding.481_of_512" 2.177979
459. "embedding.367_of_512" 2.111253
460. "embedding.383_of_512" 2.013000
461. "embedding.282_of_512" 2.006441
462. "embedding.115_of_512" 1.967250
463. "embedding.374_of_512" 1.892543
464. "embedding.055_of_512" 1.868395
465. "embedding.121_of_512" 1.431630
466. "embedding.047_of_512" 1.068393
467. "embedding.029_of_512" 1.015992
468. "embedding.031_of_512" 0.881612
469. "embedding.299_of_512" 0.785336
470. "embedding.401_of_512" 0.655422
471. "embedding.214_of_512" 0.349836
472. "embedding.248_of_512" 0.313830
473. "embedding.486_of_512" 0.227126
474. "embedding.509_of_512" 0.131360
475. "embedding.435_of_512" 0.051429
476. "embedding.034_of_512" 0.029603
Those variable importances are computed during training. More, and possibly more informative, variable importances are available when analyzing a model on a test dataset.
Num trees : 100
Only printing the first tree.
Tree #0:
"embedding.171_of_512">=-0.0184792 [s:0.0574827 n:60613 np:32412 miss:1] ; pred:1.68723e-08
├─(pos)─ "embedding.111_of_512">=-0.00413349 [s:0.0157348 n:32412 np:21552 miss:0] ; pred:-0.0906464
| ├─(pos)─ "embedding.291_of_512">=0.0227038 [s:0.00645317 n:21552 np:15994 miss:1] ; pred:-0.126738
| | ├─(pos)─ "embedding.171_of_512">=0.0260582 [s:0.00247006 n:15994 np:11830 miss:0] ; pred:-0.145932
| | | ├─(pos)─ "embedding.291_of_512">=0.070725 [s:0.00102745 n:11830 np:7037 miss:0] ; pred:-0.157884
| | | | ├─(pos)─ pred:-0.168606
| | | | └─(neg)─ pred:-0.142141
| | | └─(neg)─ "embedding.022_of_512">=-0.029009 [s:0.00552812 n:4164 np:3265 miss:1] ; pred:-0.111978
| | | ├─(pos)─ pred:-0.127791
| | | └─(neg)─ pred:-0.054546
| | └─(neg)─ "embedding.111_of_512">=0.0393142 [s:0.00708494 n:5558 np:2244 miss:0] ; pred:-0.0715036
| | ├─(pos)─ "embedding.404_of_512">=0.10271 [s:0.00525161 n:2244 np:91 miss:0] ; pred:-0.112964
| | | ├─(pos)─ pred:0.0299088
| | | └─(neg)─ pred:-0.119003
| | └─(neg)─ "embedding.171_of_512">=0.0234795 [s:0.00580773 n:3314 np:1817 miss:0] ; pred:-0.0434295
| | ├─(pos)─ pred:-0.071467
| | └─(neg)─ pred:-0.0093987
| └─(neg)─ "embedding.291_of_512">=0.0634367 [s:0.0133273 n:10860 np:4737 miss:0] ; pred:-0.0190217
| ├─(pos)─ "embedding.384_of_512">=0.0238699 [s:0.0101704 n:4737 np:966 miss:0] ; pred:-0.0722208
| | ├─(pos)─ "embedding.276_of_512">=-0.0597052 [s:0.0231348 n:966 np:722 miss:1] ; pred:0.00854195
| | | ├─(pos)─ pred:-0.0272976
| | | └─(neg)─ pred:0.114592
| | └─(neg)─ "embedding.111_of_512">=-0.0612384 [s:0.00659412 n:3771 np:3147 miss:1] ; pred:-0.0929095
| | ├─(pos)─ pred:-0.107566
| | └─(neg)─ pred:-0.0189937
| └─(neg)─ "embedding.111_of_512">=-0.0621904 [s:0.00782627 n:6123 np:5273 miss:1] ; pred:0.0221354
| ├─(pos)─ "embedding.289_of_512">=0.0119708 [s:0.00516395 n:5273 np:2436 miss:1] ; pred:0.00773873
| | ├─(pos)─ pred:-0.0236942
| | └─(neg)─ pred:0.0347287
| └─(neg)─ "embedding.140_of_512">=0.0595422 [s:0.0123933 n:850 np:263 miss:0] ; pred:0.111445
| ├─(pos)─ pred:0.0440332
| └─(neg)─ pred:0.141649
└─(neg)─ "embedding.111_of_512">=-0.047889 [s:0.0143104 n:28201 np:9723 miss:1] ; pred:0.104182
├─(pos)─ "embedding.291_of_512">=0.0518959 [s:0.0110193 n:9723 np:4373 miss:0] ; pred:0.0373387
| ├─(pos)─ "embedding.384_of_512">=0.0086953 [s:0.0114093 n:4373 np:2134 miss:0] ; pred:-0.00972295
| | ├─(pos)─ "embedding.132_of_512">=-0.0165455 [s:0.0113251 n:2134 np:1244 miss:1] ; pred:0.0346238
| | | ├─(pos)─ pred:-0.00186079
| | | └─(neg)─ pred:0.0856202
| | └─(neg)─ "embedding.368_of_512">=0.0546857 [s:0.0104556 n:2239 np:323 miss:0] ; pred:-0.05199
| | ├─(pos)─ pred:0.0489525
| | └─(neg)─ pred:-0.0690069
| └─(neg)─ "embedding.111_of_512">=0.0176729 [s:0.00952959 n:5350 np:879 miss:0] ; pred:0.0758061
| ├─(pos)─ "embedding.257_of_512">=0.0422597 [s:0.0121365 n:879 np:273 miss:0] ; pred:-0.0134314
| | ├─(pos)─ pred:-0.0799594
| | └─(neg)─ pred:0.0165391
| └─(neg)─ "embedding.171_of_512">=-0.0514695 [s:0.00725224 n:4471 np:2448 miss:1] ; pred:0.0933503
| ├─(pos)─ pred:0.0619719
| └─(neg)─ pred:0.131321
└─(neg)─ "embedding.384_of_512">=8.67305e-05 [s:0.00620107 n:18478 np:14151 miss:1] ; pred:0.139354
├─(pos)─ "embedding.111_of_512">=-0.0819011 [s:0.00399313 n:14151 np:6809 miss:1] ; pred:0.157004
| ├─(pos)─ "embedding.140_of_512">=0.0381925 [s:0.00527311 n:6809 np:1738 miss:0] ; pred:0.130407
| | ├─(pos)─ pred:0.0801316
| | └─(neg)─ pred:0.147638
| └─(neg)─ "embedding.365_of_512">=0.0224651 [s:0.00220593 n:7342 np:772 miss:0] ; pred:0.18167
| ├─(pos)─ pred:0.126134
| └─(neg)─ pred:0.188195
└─(neg)─ "embedding.022_of_512">=0.0520979 [s:0.00928281 n:4327 np:880 miss:0] ; pred:0.0816329
├─(pos)─ "embedding.483_of_512">=-0.0492134 [s:0.0154152 n:880 np:453 miss:1] ; pred:0.00434318
| ├─(pos)─ pred:0.0532019
| └─(neg)─ pred:-0.0474906
└─(neg)─ "embedding.177_of_512">=0.0136228 [s:0.00564147 n:3447 np:1675 miss:0] ; pred:0.101365
├─(pos)─ pred:0.0700516
└─(neg)─ pred:0.130963
决策森林模型使用512个数值特征作为输入。这些是由神经网络生成的512个维度的嵌入。
接下来，我们评估模型质量。
In [7]:
Copied!
decision_forest_model.evaluate(processed_test_dataset)
decision_forest_model.evaluate(processed_test_dataset)
Out[7]:
accuracy:
0.792431
AUC: '0' vs others:
0.881988
PR-AUC: '0' vs others:
0.870899
loss:
0.43559
872
872
| Label \ Pred | 1 | 0 |
|---|---|---|
| 1 | 365 | 102 |
| 0 | 79 | 326 |
为了简化生产化，通常将决策森林模型和预处理神经网络一起保存为一个单一模型是一个好方法。
In [8]:
Copied!
# 将YDF决策森林模型转换为TensorFlow函数。
tf_decision_forest_model = decision_forest_model.to_tensorflow_function()
# 将YDF决策森林模型转换为TensorFlow函数。
tf_decision_forest_model = decision_forest_model.to_tensorflow_function()
[INFO 24-04-12 10:21:03.6602 CEST kernel.cc:1233] Loading model from path /tmp/tmpicaxyl1d/ with prefix a15dba1c_ [INFO 24-04-12 10:21:03.6767 CEST quick_scorer_extended.cc:911] The binary was compiled without AVX2 support, but your CPU supports it. Enable it for faster model inference. [INFO 24-04-12 10:21:03.6774 CEST abstract_model.cc:1344] Engine "GradientBoostedTreesQuickScorerExtended" built [INFO 24-04-12 10:21:03.6775 CEST kernel.cc:1061] Use fast generic engine
In [9]:
Copied!
# 将决策森林和预处理神经网络一并保存为TensorFlow的SaveModel格式。
import tensorflow as tf
# 堆叠模型按顺序应用神经网络和决策森林模型。
class StackedModel(tf.Module):
def __init__(self, nn, df):
self._nn = nn
self._df = df
@tf.function
def __call__(self, raw_features):
sentence = raw_features["sentence"]
embedding = self._nn(sentence)
processed_features = {"embedding": embedding}
return self._df(processed_features)
stacked_model = StackedModel(pretrained_neural_network_model, tf_decision_forest_model)
# 在前3个测试样本上运行堆叠模型。
for example in raw_test_dataset.rebatch(1).take(3):
print("label:", example["label"].numpy())
print("sentence:", example["sentence"].numpy())
prediction = stacked_model({"sentence": example["sentence"]})
print("predictions:", prediction)
print("")
# 将决策森林和预处理神经网络一并保存为TensorFlow的SaveModel格式。
import tensorflow as tf
# 堆叠模型按顺序应用神经网络和决策森林模型。
class StackedModel(tf.Module):
def __init__(self, nn, df):
self._nn = nn
self._df = df
@tf.function
def __call__(self, raw_features):
sentence = raw_features["sentence"]
embedding = self._nn(sentence)
processed_features = {"embedding": embedding}
return self._df(processed_features)
stacked_model = StackedModel(pretrained_neural_network_model, tf_decision_forest_model)
# 在前3个测试样本上运行堆叠模型。
for example in raw_test_dataset.rebatch(1).take(3):
print("label:", example["label"].numpy())
print("sentence:", example["sentence"].numpy())
prediction = stacked_model({"sentence": example["sentence"]})
print("predictions:", prediction)
print("")
label: [0] sentence: [b'a valueless kiddie paean to pro basketball underwritten by the nba . '] predictions: tf.Tensor([0.8391822], shape=(1,), dtype=float32) label: [1] sentence: [b"featuring a dangerously seductive performance from the great daniel auteuil , `` sade '' covers the same period as kaufmann 's `` quills '' with more unsettlingly realistic results . "] predictions: tf.Tensor([0.23475255], shape=(1,), dtype=float32) label: [0] sentence: [b'i am sorry that i was unable to get the full brunt of the comedy . '] predictions: tf.Tensor([0.8494433], shape=(1,), dtype=float32)
2024-04-12 10:21:05.527045: W tensorflow/core/framework/local_rendezvous.cc:404] Local rendezvous is aborting with status: OUT_OF_RANGE: End of sequence
我们可以将这个堆叠模型保存以供后用。
In [10]:
Copied!
tf.saved_model.save(stacked_model, "/tmp/my_stacked_model")
tf.saved_model.save(stacked_model, "/tmp/my_stacked_model")
可以重新加载并重用该模型。
In [11]:
Copied!
loaded_stacked_model = tf.saved_model.load("/tmp/my_stacked_model")
loaded_stacked_model({"sentence":["This is a great movie"]}).numpy()
loaded_stacked_model = tf.saved_model.load("/tmp/my_stacked_model")
loaded_stacked_model({"sentence":["This is a great movie"]}).numpy()
[INFO 24-04-12 10:21:12.4042 CEST kernel.cc:1233] Loading model from path /tmp/my_stacked_model/assets/ with prefix a15dba1c_ [INFO 24-04-12 10:21:12.4235 CEST kernel.cc:1061] Use fast generic engine
Out[11]:
array([0.02682637], dtype=float32)
在上面的 "/tmp/my_stacked_model" 中保存的模型接受原始张量值作为输入。这对于Python模型开发来说是理想的。
尽管这样的模型可以在TensorFlow Serving或Vertex AI中使用，但在这些平台中，模型通常使用序列化的TensorFlow示例作为输入，并输出一个字典。以下是如何创建这样的模型：
In [12]:
Copied!
@tf.function(input_signature=[tf.TensorSpec([None], dtype=tf.string, name="inputs")])
def predict_with_with_proto_input(serialized_examples: tf.Tensor):
# 提取特征
feature_spec = {"sentence": tf.io.FixedLenFeature(shape=[], dtype=tf.string)}
features = tf.io.parse_example(serialized_examples, feature_spec)
# 做出预测
predictions = stacked_model(features)
return {"scores": predictions}
tf.saved_model.save(stacked_model, "/tmp/my_stacked_model_with_proto_input",
signatures = {tf.saved_model.DEFAULT_SERVING_SIGNATURE_DEF_KEY: predict_with_with_proto_input})
@tf.function(input_signature=[tf.TensorSpec([None], dtype=tf.string, name="inputs")])
def predict_with_with_proto_input(serialized_examples: tf.Tensor):
# 提取特征
feature_spec = {"sentence": tf.io.FixedLenFeature(shape=[], dtype=tf.string)}
features = tf.io.parse_example(serialized_examples, feature_spec)
# 做出预测
predictions = stacked_model(features)
return {"scores": predictions}
tf.saved_model.save(stacked_model, "/tmp/my_stacked_model_with_proto_input",
signatures = {tf.saved_model.DEFAULT_SERVING_SIGNATURE_DEF_KEY: predict_with_with_proto_input})
训练神经网络和决策森林模型的集成模型¶
在这个例子中，我们在一个简单的合成数据集上训练并组合一个随机森林模型、一个梯度增强树模型和一个神经网络模型。
In [13]:
Copied!
# 生成合成数据集的函数
def make_dataset(num_examples=10000, seed=1234):
np.random.seed(seed)
features = np.random.uniform(-1, 1, size=(num_examples, 4))
noise = np.random.uniform(size=(num_examples))
left_side = np.sqrt(
np.sum(np.multiply(np.square(features[:, 0:2]), [1, 2]), axis=1))
right_side = features[:, 2] * 0.7 + np.sin(
features[:, 3] * 10) * 0.5 + noise * 0.0 + 0.5
labels = left_side <= right_side
return {"features": features, "label": labels.astype(int) }
# 生成合成数据集的函数
def make_dataset(num_examples=10000, seed=1234):
np.random.seed(seed)
features = np.random.uniform(-1, 1, size=(num_examples, 4))
noise = np.random.uniform(size=(num_examples))
left_side = np.sqrt(
np.sum(np.multiply(np.square(features[:, 0:2]), [1, 2]), axis=1))
right_side = features[:, 2] * 0.7 + np.sin(
features[:, 3] * 10) * 0.5 + noise * 0.0 + 0.5
labels = left_side <= right_side
return {"features": features, "label": labels.astype(int) }
我们生成并绘制一些示例：
In [14]:
Copied!
make_dataset(num_examples=5)
make_dataset(num_examples=5)
Out[14]:
{'features': array([[-0.6169611 , 0.24421754, -0.12454452, 0.57071717],
[ 0.55995162, -0.45481479, -0.44707149, 0.60374436],
[ 0.91627871, 0.75186527, -0.28436546, 0.00199025],
[ 0.36692587, 0.42540405, -0.25949849, 0.12239237],
[ 0.00616633, -0.9724631 , 0.54565324, 0.76528238]]),
'label': array([0, 0, 0, 1, 0])}
In [15]:
Copied!
import matplotlib.pyplot as plt
plot_dataset = make_dataset(num_examples=50000)
plot_label = plot_dataset["label"]
plot_features = plot_dataset["features"]
plt.rcParams["figure.figsize"] = [8, 8]
common_args = dict(c=plot_label, s=1.0, alpha=0.5)
plt.subplot(2, 2, 1)
plt.scatter(plot_features[:, 0], plot_features[:, 1], **common_args)
plt.subplot(2, 2, 2)
plt.scatter(plot_features[:, 1], plot_features[:, 2], **common_args)
plt.subplot(2, 2, 3)
plt.scatter(plot_features[:, 0], plot_features[:, 2], **common_args)
plt.subplot(2, 2, 4)
plt.scatter(plot_features[:, 0], plot_features[:, 3], **common_args)
import matplotlib.pyplot as plt
plot_dataset = make_dataset(num_examples=50000)
plot_label = plot_dataset["label"]
plot_features = plot_dataset["features"]
plt.rcParams["figure.figsize"] = [8, 8]
common_args = dict(c=plot_label, s=1.0, alpha=0.5)
plt.subplot(2, 2, 1)
plt.scatter(plot_features[:, 0], plot_features[:, 1], **common_args)
plt.subplot(2, 2, 2)
plt.scatter(plot_features[:, 1], plot_features[:, 2], **common_args)
plt.subplot(2, 2, 3)
plt.scatter(plot_features[:, 0], plot_features[:, 2], **common_args)
plt.subplot(2, 2, 4)
plt.scatter(plot_features[:, 0], plot_features[:, 3], **common_args)
Out[15]:
<matplotlib.collections.PathCollection at 0x7f1928582e90>
注意到这个模式是平滑的，并且没有水平或垂直对齐。这将有利于神经网络模型。因为对于神经网络而言，形成圆形和非对齐的决策边界比对于决策树要容易得多。
另一方面，我们将在一个包含2500个样本的小数据集上训练模型。这将有利于决策森林模型。这是因为决策森林在使用所有可用的信息方面更加高效（决策森林是“样本高效”的）。
我们的神经网络和决策森林的集成将利用两者的优势。
In [16]:
Copied!
train_dataset = make_dataset(num_examples=10000, seed=1234)
test_dataset = make_dataset(num_examples=10000, seed=5678)
train_dataset = make_dataset(num_examples=10000, seed=1234)
test_dataset = make_dataset(num_examples=10000, seed=5678)
首先，让我们训练随机森林和梯度提升树模型。
In [17]:
Copied!
model_random_forest = ydf.RandomForestLearner(label="label").train(train_dataset)
model_gradient_boosted_trees = ydf.GradientBoostedTreesLearner(label="label").train(train_dataset)
model_random_forest = ydf.RandomForestLearner(label="label").train(train_dataset)
model_gradient_boosted_trees = ydf.GradientBoostedTreesLearner(label="label").train(train_dataset)
Train model on 10000 examples Model trained in 0:00:00.280004 Train model on 10000 examples Model trained in 0:00:01.720190
然后，让我们训练神经网络模型。
In [18]:
Copied!
def make_tf_dataset(dataset, batch_size):
# 注意：Keras 期望标签和特征是分开传递的。
return tf.data.Dataset.from_tensor_slices((
{"features":dataset["features"]},
dataset["label"],
)).batch(batch_size)
tf_train_dataset = make_tf_dataset(train_dataset, 100)
tf_test_dataset = make_tf_dataset(test_dataset, 100)
inputs = [tf.keras.Input(shape=(4,), name="features")]
x = tf.keras.layers.concatenate(inputs)
x = tf.keras.layers.Dense(10, activation="relu")(x)
x = tf.keras.layers.Dense(1)(x)
model_neural_network = tf.keras.Model(inputs=inputs, outputs=x)
model_neural_network.compile(loss=tf.keras.losses.BinaryCrossentropy(from_logits=True), metrics=["accuracy"])
model_neural_network.fit(tf_train_dataset, epochs=5)
def make_tf_dataset(dataset, batch_size):
# 注意：Keras 期望标签和特征是分开传递的。
return tf.data.Dataset.from_tensor_slices((
{"features":dataset["features"]},
dataset["label"],
)).batch(batch_size)
tf_train_dataset = make_tf_dataset(train_dataset, 100)
tf_test_dataset = make_tf_dataset(test_dataset, 100)
inputs = [tf.keras.Input(shape=(4,), name="features")]
x = tf.keras.layers.concatenate(inputs)
x = tf.keras.layers.Dense(10, activation="relu")(x)
x = tf.keras.layers.Dense(1)(x)
model_neural_network = tf.keras.Model(inputs=inputs, outputs=x)
model_neural_network.compile(loss=tf.keras.losses.BinaryCrossentropy(from_logits=True), metrics=["accuracy"])
model_neural_network.fit(tf_train_dataset, epochs=5)
Epoch 1/5 100/100 ━━━━━━━━━━━━━━━━━━━━ 0s 482us/step - accuracy: 0.7168 - loss: 0.6956 Epoch 2/5 100/100 ━━━━━━━━━━━━━━━━━━━━ 0s 412us/step - accuracy: 0.7457 - loss: 0.5865 Epoch 3/5 100/100 ━━━━━━━━━━━━━━━━━━━━ 0s 409us/step - accuracy: 0.7457 - loss: 0.5270 Epoch 4/5 100/100 ━━━━━━━━━━━━━━━━━━━━ 0s 417us/step - accuracy: 0.7457 - loss: 0.4865 Epoch 5/5 100/100 ━━━━━━━━━━━━━━━━━━━━ 0s 431us/step - accuracy: 0.7492 - loss: 0.4557
Out[18]:
<keras.src.callbacks.history.History at 0x7f1928376f50>
我们单独评估所有模型。
In [19]:
Copied!
model_random_forest.evaluate(test_dataset).accuracy
model_random_forest.evaluate(test_dataset).accuracy
Out[19]:
0.9476
In [20]:
Copied!
model_gradient_boosted_trees.evaluate(test_dataset).accuracy
model_gradient_boosted_trees.evaluate(test_dataset).accuracy
Out[20]:
0.9674
In [21]:
Copied!
model_neural_network.evaluate(tf_test_dataset, return_dict=True)["accuracy"]
model_neural_network.evaluate(tf_test_dataset, return_dict=True)["accuracy"]
100/100 ━━━━━━━━━━━━━━━━━━━━ 0s 376us/step - accuracy: 0.7546 - loss: 0.4397
Out[21]:
0.7534999847412109
让我们将树模型结合在一起。
In [22]:
Copied!
class EnsembleModel(tf.keras.Model):
def __init__(self, rf, gbt, nn):
super().__init__()
self._rf = rf
self._gbt = gbt
self._nn = nn
def call(self, features):
individual_predictions = []
# 注意：对于二分类问题，YDF输出的形状
# 是 [num example]，而 Keras 模型输出的形状
# 如果 [num example, 1]。Keras 的评估需要一个 [num example, 1]
# 成形输出。
individual_predictions.append(tf.expand_dims(self._rf(features), axis=1))
individual_predictions.append(tf.expand_dims(self._gbt(features), axis=1))
# 注意：模型输出的是一个对数几率，而不是概率。
individual_predictions.append(tf.nn.sigmoid(self._nn(features)))
return tf.reduce_mean(tf.stack(individual_predictions, axis=0), axis=0)
ensemble_model = EnsembleModel(
rf = model_random_forest.to_tensorflow_function(),
gbt = model_gradient_boosted_trees.to_tensorflow_function(),
nn = model_neural_network
)
class EnsembleModel(tf.keras.Model):
def __init__(self, rf, gbt, nn):
super().__init__()
self._rf = rf
self._gbt = gbt
self._nn = nn
def call(self, features):
individual_predictions = []
# 注意：对于二分类问题，YDF输出的形状
# 是 [num example]，而 Keras 模型输出的形状
# 如果 [num example, 1]。Keras 的评估需要一个 [num example, 1]
# 成形输出。
individual_predictions.append(tf.expand_dims(self._rf(features), axis=1))
individual_predictions.append(tf.expand_dims(self._gbt(features), axis=1))
# 注意：模型输出的是一个对数几率，而不是概率。
individual_predictions.append(tf.nn.sigmoid(self._nn(features)))
return tf.reduce_mean(tf.stack(individual_predictions, axis=0), axis=0)
ensemble_model = EnsembleModel(
rf = model_random_forest.to_tensorflow_function(),
gbt = model_gradient_boosted_trees.to_tensorflow_function(),
nn = model_neural_network
)
[INFO 24-04-12 10:21:22.5101 CEST kernel.cc:1233] Loading model from path /tmp/tmpesx684rq/ with prefix 38f4c229_ [INFO 24-04-12 10:21:23.0956 CEST decision_forest.cc:734] Model loaded with 300 root(s), 197576 node(s), and 4 input feature(s). [INFO 24-04-12 10:21:23.0956 CEST abstract_model.cc:1344] Engine "RandomForestOptPred" built [INFO 24-04-12 10:21:23.0956 CEST kernel.cc:1061] Use fast generic engine [INFO 24-04-12 10:21:23.1543 CEST kernel.cc:1233] Loading model from path /tmp/tmps0kyvz4d/ with prefix cca7e657_ [INFO 24-04-12 10:21:23.1960 CEST kernel.cc:1061] Use fast generic engine
我们打印前3个测试示例的预测结果。
In [23]:
Copied!
for features, labels in tf_test_dataset.rebatch(1).take(3):
print("features:", features)
print("labels:", labels)
prediction = ensemble_model(features)
print("prediction:", prediction)
for features, labels in tf_test_dataset.rebatch(1).take(3):
print("features:", features)
print("labels:", labels)
prediction = ensemble_model(features)
print("prediction:", prediction)
features: {'features': <tf.Tensor: shape=(1, 4), dtype=float64, numpy=array([[-0.02134604, -0.88133511, -0.26759515, 0.03773088]])>}
labels: tf.Tensor([0], shape=(1,), dtype=int64)
prediction: tf.Tensor([[0.06539446]], shape=(1, 1), dtype=float32)
features: {'features': <tf.Tensor: shape=(1, 4), dtype=float64, numpy=array([[ 0.19645002, -0.13877101, -0.64273189, -0.4294922 ]])>}
labels: tf.Tensor([1], shape=(1,), dtype=int64)
prediction: tf.Tensor([[0.57615477]], shape=(1, 1), dtype=float32)
features: {'features': <tf.Tensor: shape=(1, 4), dtype=float64, numpy=array([[-0.85712274, -0.63057332, -0.82434415, 0.47293716]])>}
labels: tf.Tensor([0], shape=(1,), dtype=int64)
prediction: tf.Tensor([[0.05497471]], shape=(1, 1), dtype=float32)
2024-04-12 10:21:23.264820: W tensorflow/core/framework/local_rendezvous.cc:404] Local rendezvous is aborting with status: OUT_OF_RANGE: End of sequence
我们评估集成模型的质量。
In [24]:
Copied!
ensemble_model.compile(loss=tf.keras.losses.CategoricalCrossentropy(), metrics=["accuracy"])
ensemble_model.evaluate(tf_test_dataset, return_dict=True)["accuracy"]
ensemble_model.compile(loss=tf.keras.losses.CategoricalCrossentropy(), metrics=["accuracy"])
ensemble_model.evaluate(tf_test_dataset, return_dict=True)["accuracy"]
25/100 ━━━━━━━━━━━━━━━━━━━━ 0s 2ms/step - accuracy: 0.9542 - loss: 3.0788e-08
/usr/local/google/home/gbm/my_venv/lib/python3.11/site-packages/keras/src/losses/losses.py:22: SyntaxWarning: In loss categorical_crossentropy, expected y_pred.shape to be (batch_size, num_classes) with num_classes > 1. Received: y_pred.shape=(None, 1). Consider using 'binary_crossentropy' if you only have 2 classes. return self.fn(y_true, y_pred, **self._fn_kwargs)
100/100 ━━━━━━━━━━━━━━━━━━━━ 0s 2ms/step - accuracy: 0.9545 - loss: 3.0870e-08
Out[24]:
0.9556999802589417
在这种情况下，集成模型的表现优于随机森林和神经网络，但在性能方面却不及梯度增强树模型。当一个模型的质量较低（在此案例中神经网络的情况），集成的整体表现可能会受到负面影响。下一步要么集中精力提高神经网络的质量，要么完全将其排除在集成之外。