pip install ydf tensorflow_hub tensorflow_datasets -U
import ydf # To train the model
import tensorflow_datasets # To download the movie review dataset
import tensorflow_hub # To download the pre-trained embedding
What is a pre-trained embedding?¶
Pretrained embeddings are models trained on a large corpus of data that can be used to improve the quality of your model when you do not have a lot of training data. Unlike a model that is trained for a specific task and outputs predictions for that task, a pretrained embedding model outputs "embeddings," which are fixed-size numerical vectors that can be used as input features for a second model (e.g. a ydf model) to solve a variety of tasks. Pre-trained embeddings are also useful for applying a model to complex or unstructured data. For example, with an image, text, audio, or video pre-trained embedding, you can apply a YDF model to image, text, audio, and video data, respectively.
In this notebook, we will classify movie reviews as either "positive" or "negative". For instance, the review beginning with "This is the kind of film for a snowy Sunday afternoon when the rest of the world can go ahead with its own business as you descend into a big arm-chair and mellow for a couple of hours. Wonderful performances from Cher and Nicolas ..." is a positive review. Our dataset contains 25000 reviews, but because 25000 reviews are NOT enough to train a good text model, and because configuring a text model is complicated, we will simply use the Universal Sentence Encoder pre-trained embedding.
Downloading dataset¶
We download the dataset from the TensorFlow Dataset repository.
raw_train_ds = tensorflow_datasets.load(name="imdb_reviews", split="train")
raw_test_ds = tensorflow_datasets.load(name="imdb_reviews", split="test")
Let's look at the first 200 letters or the first 3 examples:
for example in raw_train_ds.take(3):
print(f"""\
text: {example['text'].numpy()[:200]}
label: {example['label']}
=========================""")
text: b"This was an absolutely terrible movie. Don't be lured in by Christopher Walken or Michael Ironside. Both are great actors, but this must simply be their worst role in history. Even their great acting " label: 0 ========================= text: b'I have been known to fall asleep during films, but this is usually due to a combination of things including, really tired, being warm and comfortable on the sette and having just eaten a lot. However ' label: 0 ========================= text: b'Mann photographs the Alberta Rocky Mountains in a superb fashion, and Jimmy Stewart and Walter Brennan give enjoyable performances as they always seem to do. <br /><br />But come on Hollywood - a Moun' label: 0 =========================
Downloading embedding¶
embed = tensorflow_hub.load("https://tfhub.dev/google/universal-sentence-encoder/4")
We can test the embedding on any text. It returns a vector of numbers. While those values do not have inherent meaning to us, YDF is very good at consuming them.
embeddings = embed([
"The little blue dog eats a piece of ham.",
"It is raining today."]).numpy()
print(embeddings)
[[-0.01440776 0.04751815 0.05348268 ... 0.018609 0.03508667 -0.03631601] [-0.0552231 -0.02168638 0.05072879 ... 0.03051734 -0.00266217 0.01246582]]
Apply embedding on dataset¶
We can apply the embedding to our dataset. Since the dataset and the embedding are both created with TensorFlow, we will prepare a TensorFlow Dataset and feed it directly into YDF. YDF natively consumes TensorFlow Datasets.
def apply_embedding(batch):
batch["text"] = embed(batch["text"])
return batch
# The batch-size (256) has not impact on the YDF model. However,
# reading a TensorFlow dataset with a small (<50) batch size might
# be slow. Use a large batch size increases memory usage.
train_ds = raw_train_ds.batch(256).map(apply_embedding)
test_ds = raw_test_ds.batch(256).map(apply_embedding)
Let's show the first 10 dimensions of the embedding for the 3 examples in the first batch examples.
for example in train_ds.take(1):
print(f"""\
text: {example['text'].numpy()[:3, :10]}
label: {example['label'].numpy()[:3]}
=========================""")
text: [[ 0.04070191 0.00420414 -0.01570062 0.06623042 0.06024029 0.00345815 -0.00204514 -0.02974475 -0.06150667 0.02128238] [ 0.02308333 0.03450448 0.03191734 0.01053793 -0.004009 0.00847429 -0.03853702 -0.02518811 0.03465953 0.08872268] [ 0.0223658 -0.00636589 0.04310491 -0.05726858 0.05173567 -0.02762526 -0.0447326 -0.00299736 -0.0420398 -0.01994686]] label: [0 0 0] =========================
Training a pre-trained embedding model¶
model = ydf.GradientBoostedTreesLearner(label="label").train(train_ds)
Train model on 25000 examples Model trained in 0:00:27.819662
We can observe the 512 dimensions of the embedding. In the "variable importance" tab, we see that not all dimensions of the embedding are equally useful. For example, the feature text.111_of_512
is very useful for the model.
model.describe()
Task : CLASSIFICATION
Label : label
Features (512) : text.000_of_512 text.001_of_512 text.002_of_512 text.003_of_512 text.004_of_512 text.005_of_512 text.006_of_512 text.007_of_512 text.008_of_512 text.009_of_512 text.010_of_512 text.011_of_512 text.012_of_512 text.013_of_512 text.014_of_512 text.015_of_512 text.016_of_512 text.017_of_512 text.018_of_512 text.019_of_512 text.020_of_512 text.021_of_512 text.022_of_512 text.023_of_512 text.024_of_512 text.025_of_512 text.026_of_512 text.027_of_512 text.028_of_512 text.029_of_512 text.030_of_512 text.031_of_512 text.032_of_512 text.033_of_512 text.034_of_512 text.035_of_512 text.036_of_512 text.037_of_512 text.038_of_512 text.039_of_512 text.040_of_512 text.041_of_512 text.042_of_512 text.043_of_512 text.044_of_512 text.045_of_512 text.046_of_512 text.047_of_512 text.048_of_512 text.049_of_512 text.050_of_512 text.051_of_512 text.052_of_512 text.053_of_512 text.054_of_512 text.055_of_512 text.056_of_512 text.057_of_512 text.058_of_512 text.059_of_512 text.060_of_512 text.061_of_512 text.062_of_512 text.063_of_512 text.064_of_512 text.065_of_512 text.066_of_512 text.067_of_512 text.068_of_512 text.069_of_512 text.070_of_512 text.071_of_512 text.072_of_512 text.073_of_512 text.074_of_512 text.075_of_512 text.076_of_512 text.077_of_512 text.078_of_512 text.079_of_512 text.080_of_512 text.081_of_512 text.082_of_512 text.083_of_512 text.084_of_512 text.085_of_512 text.086_of_512 text.087_of_512 text.088_of_512 text.089_of_512 text.090_of_512 text.091_of_512 text.092_of_512 text.093_of_512 text.094_of_512 text.095_of_512 text.096_of_512 text.097_of_512 text.098_of_512 text.099_of_512 text.100_of_512 text.101_of_512 text.102_of_512 text.103_of_512 text.104_of_512 text.105_of_512 text.106_of_512 text.107_of_512 text.108_of_512 text.109_of_512 text.110_of_512 text.111_of_512 text.112_of_512 text.113_of_512 text.114_of_512 text.115_of_512 text.116_of_512 text.117_of_512 text.118_of_512 text.119_of_512 text.120_of_512 text.121_of_512 text.122_of_512 text.123_of_512 text.124_of_512 text.125_of_512 text.126_of_512 text.127_of_512 text.128_of_512 text.129_of_512 text.130_of_512 text.131_of_512 text.132_of_512 text.133_of_512 text.134_of_512 text.135_of_512 text.136_of_512 text.137_of_512 text.138_of_512 text.139_of_512 text.140_of_512 text.141_of_512 text.142_of_512 text.143_of_512 text.144_of_512 text.145_of_512 text.146_of_512 text.147_of_512 text.148_of_512 text.149_of_512 text.150_of_512 text.151_of_512 text.152_of_512 text.153_of_512 text.154_of_512 text.155_of_512 text.156_of_512 text.157_of_512 text.158_of_512 text.159_of_512 text.160_of_512 text.161_of_512 text.162_of_512 text.163_of_512 text.164_of_512 text.165_of_512 text.166_of_512 text.167_of_512 text.168_of_512 text.169_of_512 text.170_of_512 text.171_of_512 text.172_of_512 text.173_of_512 text.174_of_512 text.175_of_512 text.176_of_512 text.177_of_512 text.178_of_512 text.179_of_512 text.180_of_512 text.181_of_512 text.182_of_512 text.183_of_512 text.184_of_512 text.185_of_512 text.186_of_512 text.187_of_512 text.188_of_512 text.189_of_512 text.190_of_512 text.191_of_512 text.192_of_512 text.193_of_512 text.194_of_512 text.195_of_512 text.196_of_512 text.197_of_512 text.198_of_512 text.199_of_512 text.200_of_512 text.201_of_512 text.202_of_512 text.203_of_512 text.204_of_512 text.205_of_512 text.206_of_512 text.207_of_512 text.208_of_512 text.209_of_512 text.210_of_512 text.211_of_512 text.212_of_512 text.213_of_512 text.214_of_512 text.215_of_512 text.216_of_512 text.217_of_512 text.218_of_512 text.219_of_512 text.220_of_512 text.221_of_512 text.222_of_512 text.223_of_512 text.224_of_512 text.225_of_512 text.226_of_512 text.227_of_512 text.228_of_512 text.229_of_512 text.230_of_512 text.231_of_512 text.232_of_512 text.233_of_512 text.234_of_512 text.235_of_512 text.236_of_512 text.237_of_512 text.238_of_512 text.239_of_512 text.240_of_512 text.241_of_512 text.242_of_512 text.243_of_512 text.244_of_512 text.245_of_512 text.246_of_512 text.247_of_512 text.248_of_512 text.249_of_512 text.250_of_512 text.251_of_512 text.252_of_512 text.253_of_512 text.254_of_512 text.255_of_512 text.256_of_512 text.257_of_512 text.258_of_512 text.259_of_512 text.260_of_512 text.261_of_512 text.262_of_512 text.263_of_512 text.264_of_512 text.265_of_512 text.266_of_512 text.267_of_512 text.268_of_512 text.269_of_512 text.270_of_512 text.271_of_512 text.272_of_512 text.273_of_512 text.274_of_512 text.275_of_512 text.276_of_512 text.277_of_512 text.278_of_512 text.279_of_512 text.280_of_512 text.281_of_512 text.282_of_512 text.283_of_512 text.284_of_512 text.285_of_512 text.286_of_512 text.287_of_512 text.288_of_512 text.289_of_512 text.290_of_512 text.291_of_512 text.292_of_512 text.293_of_512 text.294_of_512 text.295_of_512 text.296_of_512 text.297_of_512 text.298_of_512 text.299_of_512 text.300_of_512 text.301_of_512 text.302_of_512 text.303_of_512 text.304_of_512 text.305_of_512 text.306_of_512 text.307_of_512 text.308_of_512 text.309_of_512 text.310_of_512 text.311_of_512 text.312_of_512 text.313_of_512 text.314_of_512 text.315_of_512 text.316_of_512 text.317_of_512 text.318_of_512 text.319_of_512 text.320_of_512 text.321_of_512 text.322_of_512 text.323_of_512 text.324_of_512 text.325_of_512 text.326_of_512 text.327_of_512 text.328_of_512 text.329_of_512 text.330_of_512 text.331_of_512 text.332_of_512 text.333_of_512 text.334_of_512 text.335_of_512 text.336_of_512 text.337_of_512 text.338_of_512 text.339_of_512 text.340_of_512 text.341_of_512 text.342_of_512 text.343_of_512 text.344_of_512 text.345_of_512 text.346_of_512 text.347_of_512 text.348_of_512 text.349_of_512 text.350_of_512 text.351_of_512 text.352_of_512 text.353_of_512 text.354_of_512 text.355_of_512 text.356_of_512 text.357_of_512 text.358_of_512 text.359_of_512 text.360_of_512 text.361_of_512 text.362_of_512 text.363_of_512 text.364_of_512 text.365_of_512 text.366_of_512 text.367_of_512 text.368_of_512 text.369_of_512 text.370_of_512 text.371_of_512 text.372_of_512 text.373_of_512 text.374_of_512 text.375_of_512 text.376_of_512 text.377_of_512 text.378_of_512 text.379_of_512 text.380_of_512 text.381_of_512 text.382_of_512 text.383_of_512 text.384_of_512 text.385_of_512 text.386_of_512 text.387_of_512 text.388_of_512 text.389_of_512 text.390_of_512 text.391_of_512 text.392_of_512 text.393_of_512 text.394_of_512 text.395_of_512 text.396_of_512 text.397_of_512 text.398_of_512 text.399_of_512 text.400_of_512 text.401_of_512 text.402_of_512 text.403_of_512 text.404_of_512 text.405_of_512 text.406_of_512 text.407_of_512 text.408_of_512 text.409_of_512 text.410_of_512 text.411_of_512 text.412_of_512 text.413_of_512 text.414_of_512 text.415_of_512 text.416_of_512 text.417_of_512 text.418_of_512 text.419_of_512 text.420_of_512 text.421_of_512 text.422_of_512 text.423_of_512 text.424_of_512 text.425_of_512 text.426_of_512 text.427_of_512 text.428_of_512 text.429_of_512 text.430_of_512 text.431_of_512 text.432_of_512 text.433_of_512 text.434_of_512 text.435_of_512 text.436_of_512 text.437_of_512 text.438_of_512 text.439_of_512 text.440_of_512 text.441_of_512 text.442_of_512 text.443_of_512 text.444_of_512 text.445_of_512 text.446_of_512 text.447_of_512 text.448_of_512 text.449_of_512 text.450_of_512 text.451_of_512 text.452_of_512 text.453_of_512 text.454_of_512 text.455_of_512 text.456_of_512 text.457_of_512 text.458_of_512 text.459_of_512 text.460_of_512 text.461_of_512 text.462_of_512 text.463_of_512 text.464_of_512 text.465_of_512 text.466_of_512 text.467_of_512 text.468_of_512 text.469_of_512 text.470_of_512 text.471_of_512 text.472_of_512 text.473_of_512 text.474_of_512 text.475_of_512 text.476_of_512 text.477_of_512 text.478_of_512 text.479_of_512 text.480_of_512 text.481_of_512 text.482_of_512 text.483_of_512 text.484_of_512 text.485_of_512 text.486_of_512 text.487_of_512 text.488_of_512 text.489_of_512 text.490_of_512 text.491_of_512 text.492_of_512 text.493_of_512 text.494_of_512 text.495_of_512 text.496_of_512 text.497_of_512 text.498_of_512 text.499_of_512 text.500_of_512 text.501_of_512 text.502_of_512 text.503_of_512 text.504_of_512 text.505_of_512 text.506_of_512 text.507_of_512 text.508_of_512 text.509_of_512 text.510_of_512 text.511_of_512
Weights : None
Trained with tuner : No
Model size : 4220 kB
Number of records: 25000 Number of columns: 513 Number of columns by type: NUMERICAL: 512 (99.8051%) CATEGORICAL: 1 (0.194932%) Columns: NUMERICAL: 512 (99.8051%) 1: "text.000_of_512" NUMERICAL mean:-0.0116812 min:-0.0802732 max:0.0762275 sd:0.0288812 dtype:DTYPE_FLOAT32 2: "text.001_of_512" NUMERICAL mean:-0.0138848 min:-0.0922933 max:0.0878271 sd:0.0356528 dtype:DTYPE_FLOAT32 3: "text.002_of_512" NUMERICAL mean:0.00793811 min:-0.0816057 max:0.104839 sd:0.0379118 dtype:DTYPE_FLOAT32 4: "text.003_of_512" NUMERICAL mean:0.0173988 min:-0.086644 max:0.0888103 sd:0.0356921 dtype:DTYPE_FLOAT32 5: "text.004_of_512" NUMERICAL mean:0.0223057 min:-0.0912103 max:0.0840179 sd:0.0366385 dtype:DTYPE_FLOAT32 6: "text.005_of_512" NUMERICAL mean:0.0112871 min:-0.0888469 max:0.0951167 sd:0.0403517 dtype:DTYPE_FLOAT32 7: "text.006_of_512" NUMERICAL mean:0.00491578 min:-0.0835639 max:0.0810243 sd:0.0361497 dtype:DTYPE_FLOAT32 8: "text.007_of_512" NUMERICAL mean:0.00392599 min:-0.0818771 max:0.101976 sd:0.0377445 dtype:DTYPE_FLOAT32 9: "text.008_of_512" NUMERICAL mean:0.0309414 min:-0.11085 max:0.105491 sd:0.0450593 dtype:DTYPE_FLOAT32 10: "text.009_of_512" NUMERICAL mean:0.00750527 min:-0.0955643 max:0.0887227 sd:0.0378264 dtype:DTYPE_FLOAT32 11: "text.010_of_512" NUMERICAL mean:0.0441103 min:-0.0798154 max:0.0977228 sd:0.031588 dtype:DTYPE_FLOAT32 12: "text.011_of_512" NUMERICAL mean:0.00263192 min:-0.0852525 max:0.0870859 sd:0.03544 dtype:DTYPE_FLOAT32 13: "text.012_of_512" NUMERICAL mean:0.00863618 min:-0.0815591 max:0.0835928 sd:0.0342734 dtype:DTYPE_FLOAT32 14: "text.013_of_512" NUMERICAL mean:0.0050239 min:-0.0799308 max:0.102996 sd:0.0388532 dtype:DTYPE_FLOAT32 15: "text.014_of_512" NUMERICAL mean:0.0230528 min:-0.0888987 max:0.0865913 sd:0.0346821 dtype:DTYPE_FLOAT32 16: "text.015_of_512" NUMERICAL mean:-0.0606065 min:-0.104098 max:0.0835962 sd:0.0218493 dtype:DTYPE_FLOAT32 17: "text.016_of_512" NUMERICAL mean:-0.0145868 min:-0.101336 max:0.0818689 sd:0.0397594 dtype:DTYPE_FLOAT32 18: "text.017_of_512" NUMERICAL mean:0.00628277 min:-0.0896906 max:0.0844325 sd:0.0405856 dtype:DTYPE_FLOAT32 19: "text.018_of_512" NUMERICAL mean:0.0242312 min:-0.0832288 max:0.094017 sd:0.0382305 dtype:DTYPE_FLOAT32 20: "text.019_of_512" NUMERICAL mean:0.00299067 min:-0.101912 max:0.083286 sd:0.0387058 dtype:DTYPE_FLOAT32 21: "text.020_of_512" NUMERICAL mean:0.00716193 min:-0.0782654 max:0.0897766 sd:0.0345465 dtype:DTYPE_FLOAT32 22: "text.021_of_512" NUMERICAL mean:-0.0645249 min:-0.116693 max:0.0741038 sd:0.0279506 dtype:DTYPE_FLOAT32 23: "text.022_of_512" NUMERICAL mean:0.00219255 min:-0.0849867 max:0.0963482 sd:0.0446306 dtype:DTYPE_FLOAT32 24: "text.023_of_512" NUMERICAL mean:-0.00826867 min:-0.0830943 max:0.0915099 sd:0.0385256 dtype:DTYPE_FLOAT32 25: "text.024_of_512" NUMERICAL mean:0.0176837 min:-0.109484 max:0.0887832 sd:0.045418 dtype:DTYPE_FLOAT32 26: "text.025_of_512" NUMERICAL mean:0.0187665 min:-0.0900682 max:0.0838608 sd:0.0353671 dtype:DTYPE_FLOAT32 27: "text.026_of_512" NUMERICAL mean:-0.0215764 min:-0.0890075 max:0.0778969 sd:0.0340446 dtype:DTYPE_FLOAT32 28: "text.027_of_512" NUMERICAL mean:-0.00446867 min:-0.0866572 max:0.0898454 sd:0.0358309 dtype:DTYPE_FLOAT32 29: "text.028_of_512" NUMERICAL mean:0.00127866 min:-0.0980607 max:0.0828708 sd:0.0406601 dtype:DTYPE_FLOAT32 30: "text.029_of_512" NUMERICAL mean:0.0635234 min:-0.0871338 max:0.119664 sd:0.0330037 dtype:DTYPE_FLOAT32 31: "text.030_of_512" NUMERICAL mean:0.0193726 min:-0.0782317 max:0.0890966 sd:0.0353165 dtype:DTYPE_FLOAT32 32: "text.031_of_512" NUMERICAL mean:6.14554e-05 min:-0.0889933 max:0.0959262 sd:0.0370509 dtype:DTYPE_FLOAT32 33: "text.032_of_512" NUMERICAL mean:-0.000461868 min:-0.0839903 max:0.0898939 sd:0.0346664 dtype:DTYPE_FLOAT32 34: "text.033_of_512" NUMERICAL mean:-0.014042 min:-0.0912057 max:0.0867956 sd:0.0358765 dtype:DTYPE_FLOAT32 35: "text.034_of_512" NUMERICAL mean:-0.0044079 min:-0.0867305 max:0.0900552 sd:0.036498 dtype:DTYPE_FLOAT32 36: "text.035_of_512" NUMERICAL mean:0.00279672 min:-0.0842639 max:0.0910702 sd:0.0383746 dtype:DTYPE_FLOAT32 37: "text.036_of_512" NUMERICAL mean:0.0162916 min:-0.0866203 max:0.0926366 sd:0.0364715 dtype:DTYPE_FLOAT32 38: "text.037_of_512" NUMERICAL mean:0.00715524 min:-0.0873235 max:0.0946155 sd:0.034479 dtype:DTYPE_FLOAT32 39: "text.038_of_512" NUMERICAL mean:-0.014824 min:-0.0857881 max:0.0991842 sd:0.0367527 dtype:DTYPE_FLOAT32 40: "text.039_of_512" NUMERICAL mean:-0.0119438 min:-0.0839529 max:0.0807069 sd:0.0352535 dtype:DTYPE_FLOAT32 41: "text.040_of_512" NUMERICAL mean:-0.000874137 min:-0.0808055 max:0.099789 sd:0.0361416 dtype:DTYPE_FLOAT32 42: "text.041_of_512" NUMERICAL mean:-0.0196322 min:-0.0980138 max:0.0799274 sd:0.0355608 dtype:DTYPE_FLOAT32 43: "text.042_of_512" NUMERICAL mean:-0.0491474 min:-0.11458 max:0.0973998 sd:0.0427591 dtype:DTYPE_FLOAT32 44: "text.043_of_512" NUMERICAL mean:-0.0116106 min:-0.0876132 max:0.0968711 sd:0.0310738 dtype:DTYPE_FLOAT32 45: "text.044_of_512" NUMERICAL mean:-0.00931938 min:-0.0952422 max:0.0868679 sd:0.0354728 dtype:DTYPE_FLOAT32 46: "text.045_of_512" NUMERICAL mean:-0.00600887 min:-0.0851451 max:0.0828623 sd:0.0356077 dtype:DTYPE_FLOAT32 47: "text.046_of_512" NUMERICAL mean:0.0124154 min:-0.102451 max:0.0815266 sd:0.0339426 dtype:DTYPE_FLOAT32 48: "text.047_of_512" NUMERICAL mean:-0.00657923 min:-0.0920031 max:0.0833555 sd:0.0362987 dtype:DTYPE_FLOAT32 49: "text.048_of_512" NUMERICAL mean:-0.00678232 min:-0.085266 max:0.0833177 sd:0.0376078 dtype:DTYPE_FLOAT32 50: "text.049_of_512" NUMERICAL mean:0.0160775 min:-0.0809174 max:0.0929498 sd:0.0361883 dtype:DTYPE_FLOAT32 51: "text.050_of_512" NUMERICAL mean:0.034919 min:-0.0865508 max:0.106974 sd:0.0388835 dtype:DTYPE_FLOAT32 52: "text.051_of_512" NUMERICAL mean:0.00522677 min:-0.105355 max:0.0998182 sd:0.0465767 dtype:DTYPE_FLOAT32 53: "text.052_of_512" NUMERICAL mean:0.00683111 min:-0.0794802 max:0.0934055 sd:0.0336335 dtype:DTYPE_FLOAT32 54: "text.053_of_512" NUMERICAL mean:-0.0238304 min:-0.0834709 max:0.0759645 sd:0.0314589 dtype:DTYPE_FLOAT32 55: "text.054_of_512" NUMERICAL mean:0.000111696 min:-0.0859526 max:0.0824052 sd:0.0365313 dtype:DTYPE_FLOAT32 56: "text.055_of_512" NUMERICAL mean:-0.00813263 min:-0.0991159 max:0.0898171 sd:0.0405658 dtype:DTYPE_FLOAT32 57: "text.056_of_512" NUMERICAL mean:-0.0377465 min:-0.0932685 max:0.0864 sd:0.0329359 dtype:DTYPE_FLOAT32 58: "text.057_of_512" NUMERICAL mean:0.00809272 min:-0.0926483 max:0.0829849 sd:0.0393339 dtype:DTYPE_FLOAT32 59: "text.058_of_512" NUMERICAL mean:0.0319977 min:-0.0828434 max:0.0904496 sd:0.0321783 dtype:DTYPE_FLOAT32 60: "text.059_of_512" NUMERICAL mean:-0.00708531 min:-0.0930212 max:0.0808008 sd:0.0353817 dtype:DTYPE_FLOAT32 61: "text.060_of_512" NUMERICAL mean:0.055502 min:-0.083803 max:0.111756 sd:0.0313429 dtype:DTYPE_FLOAT32 62: "text.061_of_512" NUMERICAL mean:-0.0221096 min:-0.0879145 max:0.0872049 sd:0.0363354 dtype:DTYPE_FLOAT32 63: "text.062_of_512" NUMERICAL mean:-0.0555259 min:-0.115528 max:0.086112 sd:0.0363024 dtype:DTYPE_FLOAT32 64: "text.063_of_512" NUMERICAL mean:-0.0175932 min:-0.111511 max:0.0857765 sd:0.037374 dtype:DTYPE_FLOAT32 65: "text.064_of_512" NUMERICAL mean:-0.0167136 min:-0.0847294 max:0.0872757 sd:0.034792 dtype:DTYPE_FLOAT32 66: "text.065_of_512" NUMERICAL mean:0.0292565 min:-0.0912326 max:0.0985996 sd:0.0374893 dtype:DTYPE_FLOAT32 67: "text.066_of_512" NUMERICAL mean:0.0192955 min:-0.0876318 max:0.0925277 sd:0.0350876 dtype:DTYPE_FLOAT32 68: "text.067_of_512" NUMERICAL mean:-0.00520634 min:-0.092069 max:0.0895853 sd:0.0372292 dtype:DTYPE_FLOAT32 69: "text.068_of_512" NUMERICAL mean:0.0190194 min:-0.0757692 max:0.0857291 sd:0.0333796 dtype:DTYPE_FLOAT32 70: "text.069_of_512" NUMERICAL mean:-0.000812311 min:-0.0850013 max:0.0871921 sd:0.0388029 dtype:DTYPE_FLOAT32 71: "text.070_of_512" NUMERICAL mean:0.0161897 min:-0.095693 max:0.0898743 sd:0.0370202 dtype:DTYPE_FLOAT32 72: "text.071_of_512" NUMERICAL mean:0.0630804 min:-0.0936433 max:0.0968621 sd:0.0185095 dtype:DTYPE_FLOAT32 73: "text.072_of_512" NUMERICAL mean:-0.0104304 min:-0.0882601 max:0.0839045 sd:0.0365536 dtype:DTYPE_FLOAT32 74: "text.073_of_512" NUMERICAL mean:0.0140369 min:-0.0893741 max:0.0970604 sd:0.0368163 dtype:DTYPE_FLOAT32 75: "text.074_of_512" NUMERICAL mean:-0.000649945 min:-0.0842873 max:0.0853423 sd:0.0349066 dtype:DTYPE_FLOAT32 76: "text.075_of_512" NUMERICAL mean:-0.000800347 min:-0.0918052 max:0.0882493 sd:0.0386586 dtype:DTYPE_FLOAT32 77: "text.076_of_512" NUMERICAL mean:0.00913866 min:-0.0818236 max:0.085171 sd:0.0348286 dtype:DTYPE_FLOAT32 78: "text.077_of_512" NUMERICAL mean:0.015921 min:-0.0877 max:0.0833098 sd:0.0357491 dtype:DTYPE_FLOAT32 79: "text.078_of_512" NUMERICAL mean:0.00405493 min:-0.0839781 max:0.088876 sd:0.0363744 dtype:DTYPE_FLOAT32 80: "text.079_of_512" NUMERICAL mean:0.0381824 min:-0.102282 max:0.107452 sd:0.0373116 dtype:DTYPE_FLOAT32 81: "text.080_of_512" NUMERICAL mean:-0.00555331 min:-0.0842675 max:0.0964607 sd:0.043293 dtype:DTYPE_FLOAT32 82: "text.081_of_512" NUMERICAL mean:-0.00147837 min:-0.0831993 max:0.0865752 sd:0.0365164 dtype:DTYPE_FLOAT32 83: "text.082_of_512" NUMERICAL mean:0.025426 min:-0.0937223 max:0.086701 sd:0.0257131 dtype:DTYPE_FLOAT32 84: "text.083_of_512" NUMERICAL mean:0.00398569 min:-0.0843214 max:0.0978019 sd:0.0368501 dtype:DTYPE_FLOAT32 85: "text.084_of_512" NUMERICAL mean:-0.0110377 min:-0.0803288 max:0.0817068 sd:0.0300815 dtype:DTYPE_FLOAT32 86: "text.085_of_512" NUMERICAL mean:-0.0363215 min:-0.0996178 max:0.0953932 sd:0.0364677 dtype:DTYPE_FLOAT32 87: "text.086_of_512" NUMERICAL mean:0.0233306 min:-0.0918728 max:0.0879645 sd:0.0328336 dtype:DTYPE_FLOAT32 88: "text.087_of_512" NUMERICAL mean:0.00631495 min:-0.0881925 max:0.0916896 sd:0.041526 dtype:DTYPE_FLOAT32 89: "text.088_of_512" NUMERICAL mean:-0.0627343 min:-0.10116 max:0.0783002 sd:0.0169812 dtype:DTYPE_FLOAT32 90: "text.089_of_512" NUMERICAL mean:0.00859365 min:-0.0843794 max:0.0880783 sd:0.0367226 dtype:DTYPE_FLOAT32 91: "text.090_of_512" NUMERICAL mean:-0.0148234 min:-0.0992516 max:0.0867767 sd:0.0350343 dtype:DTYPE_FLOAT32 92: "text.091_of_512" NUMERICAL mean:-0.0267766 min:-0.0942044 max:0.0793247 sd:0.0378253 dtype:DTYPE_FLOAT32 93: "text.092_of_512" NUMERICAL mean:0.00188643 min:-0.0887884 max:0.0891514 sd:0.0328687 dtype:DTYPE_FLOAT32 94: "text.093_of_512" NUMERICAL mean:-0.00867184 min:-0.0904799 max:0.0835108 sd:0.0382605 dtype:DTYPE_FLOAT32 95: "text.094_of_512" NUMERICAL mean:0.0672572 min:-0.0779515 max:0.116603 sd:0.022315 dtype:DTYPE_FLOAT32 96: "text.095_of_512" NUMERICAL mean:0.0171949 min:-0.0842528 max:0.101687 sd:0.041133 dtype:DTYPE_FLOAT32 97: "text.096_of_512" NUMERICAL mean:0.00215731 min:-0.0919103 max:0.0858262 sd:0.0326993 dtype:DTYPE_FLOAT32 98: "text.097_of_512" NUMERICAL mean:0.0519067 min:-0.0837317 max:0.0922723 sd:0.0247894 dtype:DTYPE_FLOAT32 99: "text.098_of_512" NUMERICAL mean:0.0746002 min:-0.0572844 max:0.115223 sd:0.00885761 dtype:DTYPE_FLOAT32 100: "text.099_of_512" NUMERICAL mean:0.0332451 min:-0.0907926 max:0.105707 sd:0.0386232 dtype:DTYPE_FLOAT32 101: "text.100_of_512" NUMERICAL mean:0.0109933 min:-0.0813328 max:0.0985877 sd:0.0408154 dtype:DTYPE_FLOAT32 102: "text.101_of_512" NUMERICAL mean:0.0181737 min:-0.0877092 max:0.0845753 sd:0.0328557 dtype:DTYPE_FLOAT32 103: "text.102_of_512" NUMERICAL mean:-0.00654396 min:-0.0983599 max:0.0980366 sd:0.0356183 dtype:DTYPE_FLOAT32 104: "text.103_of_512" NUMERICAL mean:-0.0220846 min:-0.0850007 max:0.0918551 sd:0.0368446 dtype:DTYPE_FLOAT32 105: "text.104_of_512" NUMERICAL mean:0.00593256 min:-0.0855101 max:0.0941251 sd:0.0359535 dtype:DTYPE_FLOAT32 106: "text.105_of_512" NUMERICAL mean:0.0191588 min:-0.084062 max:0.0861212 sd:0.0360096 dtype:DTYPE_FLOAT32 107: "text.106_of_512" NUMERICAL mean:-0.0307491 min:-0.0848395 max:0.0835213 sd:0.0319278 dtype:DTYPE_FLOAT32 108: "text.107_of_512" NUMERICAL mean:-0.0187303 min:-0.0852794 max:0.0841794 sd:0.0327629 dtype:DTYPE_FLOAT32 109: "text.108_of_512" NUMERICAL mean:-0.0200331 min:-0.0824767 max:0.0806784 sd:0.0342799 dtype:DTYPE_FLOAT32 110: "text.109_of_512" NUMERICAL mean:0.0329799 min:-0.0862293 max:0.0861934 sd:0.0281822 dtype:DTYPE_FLOAT32 111: "text.110_of_512" NUMERICAL mean:-0.0276342 min:-0.0992642 max:0.0821939 sd:0.0383224 dtype:DTYPE_FLOAT32 112: "text.111_of_512" NUMERICAL mean:-0.0385349 min:-0.125692 max:0.0870559 sd:0.0429422 dtype:DTYPE_FLOAT32 113: "text.112_of_512" NUMERICAL mean:0.0193794 min:-0.114199 max:0.0977234 sd:0.0361729 dtype:DTYPE_FLOAT32 114: "text.113_of_512" NUMERICAL mean:0.0112268 min:-0.0864682 max:0.0858257 sd:0.039243 dtype:DTYPE_FLOAT32 115: "text.114_of_512" NUMERICAL mean:0.0123548 min:-0.0952021 max:0.0832505 sd:0.0360632 dtype:DTYPE_FLOAT32 116: "text.115_of_512" NUMERICAL mean:-0.00950765 min:-0.104888 max:0.0763793 sd:0.0293429 dtype:DTYPE_FLOAT32 117: "text.116_of_512" NUMERICAL mean:-0.017355 min:-0.089618 max:0.102031 sd:0.0366594 dtype:DTYPE_FLOAT32 118: "text.117_of_512" NUMERICAL mean:-0.0167883 min:-0.0873107 max:0.0870495 sd:0.0342432 dtype:DTYPE_FLOAT32 119: "text.118_of_512" NUMERICAL mean:0.00556759 min:-0.0968067 max:0.0858974 sd:0.0407264 dtype:DTYPE_FLOAT32 120: "text.119_of_512" NUMERICAL mean:-0.0126509 min:-0.0907735 max:0.0857452 sd:0.036529 dtype:DTYPE_FLOAT32 121: "text.120_of_512" NUMERICAL mean:-0.000269181 min:-0.0875327 max:0.0926786 sd:0.0348444 dtype:DTYPE_FLOAT32 122: "text.121_of_512" NUMERICAL mean:0.0174495 min:-0.0830898 max:0.0890086 sd:0.034934 dtype:DTYPE_FLOAT32 123: "text.122_of_512" NUMERICAL mean:-0.00402577 min:-0.080368 max:0.0873479 sd:0.0360286 dtype:DTYPE_FLOAT32 124: "text.123_of_512" NUMERICAL mean:-0.00606442 min:-0.0911265 max:0.0814825 sd:0.0366954 dtype:DTYPE_FLOAT32 125: "text.124_of_512" NUMERICAL mean:-0.00149129 min:-0.0959048 max:0.0828692 sd:0.0394369 dtype:DTYPE_FLOAT32 126: "text.125_of_512" NUMERICAL mean:-0.0370789 min:-0.0979406 max:0.0694385 sd:0.0272071 dtype:DTYPE_FLOAT32 127: "text.126_of_512" NUMERICAL mean:0.00694822 min:-0.0828382 max:0.0994004 sd:0.0411108 dtype:DTYPE_FLOAT32 128: "text.127_of_512" NUMERICAL mean:-0.0758831 min:-0.121351 max:0.0521276 sd:0.00896978 dtype:DTYPE_FLOAT32 129: "text.128_of_512" NUMERICAL mean:0.0260232 min:-0.0874072 max:0.0858674 sd:0.0326943 dtype:DTYPE_FLOAT32 130: "text.129_of_512" NUMERICAL mean:0.057665 min:-0.0758189 max:0.112 sd:0.0289527 dtype:DTYPE_FLOAT32 131: "text.130_of_512" NUMERICAL mean:0.000940865 min:-0.083708 max:0.0820816 sd:0.0351123 dtype:DTYPE_FLOAT32 132: "text.131_of_512" NUMERICAL mean:-0.00187575 min:-0.0815817 max:0.091749 sd:0.0343653 dtype:DTYPE_FLOAT32 133: "text.132_of_512" NUMERICAL mean:0.03079 min:-0.0723224 max:0.0852465 sd:0.0274576 dtype:DTYPE_FLOAT32 134: "text.133_of_512" NUMERICAL mean:-0.0284279 min:-0.0903906 max:0.0851924 sd:0.0359698 dtype:DTYPE_FLOAT32 135: "text.134_of_512" NUMERICAL mean:0.0158556 min:-0.0844077 max:0.083871 sd:0.0334406 dtype:DTYPE_FLOAT32 136: "text.135_of_512" NUMERICAL mean:-0.0352063 min:-0.0955508 max:0.0792998 sd:0.032951 dtype:DTYPE_FLOAT32 137: "text.136_of_512" NUMERICAL mean:0.0241423 min:-0.084793 max:0.0811507 sd:0.0296611 dtype:DTYPE_FLOAT32 138: "text.137_of_512" NUMERICAL mean:-0.0013576 min:-0.0846433 max:0.0931272 sd:0.0379585 dtype:DTYPE_FLOAT32 139: "text.138_of_512" NUMERICAL mean:-0.0559221 min:-0.125214 max:0.0643445 sd:0.0232717 dtype:DTYPE_FLOAT32 140: "text.139_of_512" NUMERICAL mean:0.00332664 min:-0.0838819 max:0.0806912 sd:0.034472 dtype:DTYPE_FLOAT32 141: "text.140_of_512" NUMERICAL mean:0.0223662 min:-0.0847569 max:0.100437 sd:0.0444653 dtype:DTYPE_FLOAT32 142: "text.141_of_512" NUMERICAL mean:0.0731112 min:-0.0622762 max:0.115494 sd:0.00969383 dtype:DTYPE_FLOAT32 143: "text.142_of_512" NUMERICAL mean:-0.0260666 min:-0.0907967 max:0.0868847 sd:0.0383228 dtype:DTYPE_FLOAT32 144: "text.143_of_512" NUMERICAL mean:0.0480093 min:-0.0955401 max:0.100895 sd:0.0329786 dtype:DTYPE_FLOAT32 145: "text.144_of_512" NUMERICAL mean:-0.00939173 min:-0.0864651 max:0.0867805 sd:0.0363782 dtype:DTYPE_FLOAT32 146: "text.145_of_512" NUMERICAL mean:-0.0148715 min:-0.0866322 max:0.0858908 sd:0.0394167 dtype:DTYPE_FLOAT32 147: "text.146_of_512" NUMERICAL mean:0.00624596 min:-0.0876547 max:0.0766023 sd:0.0321276 dtype:DTYPE_FLOAT32 148: "text.147_of_512" NUMERICAL mean:-0.00343463 min:-0.0825002 max:0.0856828 sd:0.0363225 dtype:DTYPE_FLOAT32 149: "text.148_of_512" NUMERICAL mean:0.0268712 min:-0.0893011 max:0.0811012 sd:0.0295636 dtype:DTYPE_FLOAT32 150: "text.149_of_512" NUMERICAL mean:0.00122339 min:-0.085816 max:0.0883449 sd:0.0360746 dtype:DTYPE_FLOAT32 151: "text.150_of_512" NUMERICAL mean:-0.0105128 min:-0.0872871 max:0.0837847 sd:0.0359024 dtype:DTYPE_FLOAT32 152: "text.151_of_512" NUMERICAL mean:-0.0167417 min:-0.084347 max:0.0759722 sd:0.0297068 dtype:DTYPE_FLOAT32 153: "text.152_of_512" NUMERICAL mean:-0.0388598 min:-0.10226 max:0.0867861 sd:0.0391177 dtype:DTYPE_FLOAT32 154: "text.153_of_512" NUMERICAL mean:-0.00644222 min:-0.0985692 max:0.0797407 sd:0.0365828 dtype:DTYPE_FLOAT32 155: "text.154_of_512" NUMERICAL mean:-0.0139682 min:-0.0825483 max:0.0846248 sd:0.0347379 dtype:DTYPE_FLOAT32 156: "text.155_of_512" NUMERICAL mean:0.0368961 min:-0.0838295 max:0.0845345 sd:0.0285216 dtype:DTYPE_FLOAT32 157: "text.156_of_512" NUMERICAL mean:-0.0054991 min:-0.0867518 max:0.0852941 sd:0.0366612 dtype:DTYPE_FLOAT32 158: "text.157_of_512" NUMERICAL mean:0.0262428 min:-0.0727893 max:0.100899 sd:0.029308 dtype:DTYPE_FLOAT32 159: "text.158_of_512" NUMERICAL mean:0.0281793 min:-0.0757985 max:0.103439 sd:0.0317215 dtype:DTYPE_FLOAT32 160: "text.159_of_512" NUMERICAL mean:0.0151069 min:-0.0849506 max:0.0951084 sd:0.0425651 dtype:DTYPE_FLOAT32 161: "text.160_of_512" NUMERICAL mean:-0.0138389 min:-0.0934066 max:0.0879302 sd:0.0394525 dtype:DTYPE_FLOAT32 162: "text.161_of_512" NUMERICAL mean:-0.014178 min:-0.083802 max:0.0811372 sd:0.0343825 dtype:DTYPE_FLOAT32 163: "text.162_of_512" NUMERICAL mean:0.0194956 min:-0.0791388 max:0.0900103 sd:0.0332954 dtype:DTYPE_FLOAT32 164: "text.163_of_512" NUMERICAL mean:-0.00653962 min:-0.0935783 max:0.0882855 sd:0.0379969 dtype:DTYPE_FLOAT32 165: "text.164_of_512" NUMERICAL mean:0.0296711 min:-0.0863985 max:0.0939484 sd:0.0323785 dtype:DTYPE_FLOAT32 166: "text.165_of_512" NUMERICAL mean:0.0243118 min:-0.0818753 max:0.0847991 sd:0.0323827 dtype:DTYPE_FLOAT32 167: "text.166_of_512" NUMERICAL mean:-0.012329 min:-0.0898639 max:0.0979821 sd:0.0424687 dtype:DTYPE_FLOAT32 168: "text.167_of_512" NUMERICAL mean:-0.00509269 min:-0.0855203 max:0.0805767 sd:0.0363154 dtype:DTYPE_FLOAT32 169: "text.168_of_512" NUMERICAL mean:-0.0174308 min:-0.10872 max:0.0811491 sd:0.0335062 dtype:DTYPE_FLOAT32 170: "text.169_of_512" NUMERICAL mean:-0.00234378 min:-0.0811347 max:0.0807904 sd:0.0339118 dtype:DTYPE_FLOAT32 171: "text.170_of_512" NUMERICAL mean:0.0390872 min:-0.0884981 max:0.0905207 sd:0.0305453 dtype:DTYPE_FLOAT32 172: "text.171_of_512" NUMERICAL mean:-0.0417291 min:-0.124667 max:0.0893347 sd:0.0376908 dtype:DTYPE_FLOAT32 173: "text.172_of_512" NUMERICAL mean:-0.018656 min:-0.0907259 max:0.0872195 sd:0.0355536 dtype:DTYPE_FLOAT32 174: "text.173_of_512" NUMERICAL mean:-0.00363774 min:-0.101118 max:0.0862936 sd:0.0429395 dtype:DTYPE_FLOAT32 175: "text.174_of_512" NUMERICAL mean:-0.0181822 min:-0.0886859 max:0.0848198 sd:0.0374609 dtype:DTYPE_FLOAT32 176: "text.175_of_512" NUMERICAL mean:0.0167266 min:-0.0897007 max:0.0849533 sd:0.0333302 dtype:DTYPE_FLOAT32 177: "text.176_of_512" NUMERICAL mean:0.0231168 min:-0.0889035 max:0.0829836 sd:0.0346788 dtype:DTYPE_FLOAT32 178: "text.177_of_512" NUMERICAL mean:0.011108 min:-0.0841251 max:0.085787 sd:0.0373291 dtype:DTYPE_FLOAT32 179: "text.178_of_512" NUMERICAL mean:0.0265751 min:-0.0921168 max:0.0870286 sd:0.0364188 dtype:DTYPE_FLOAT32 180: "text.179_of_512" NUMERICAL mean:-0.0653105 min:-0.105651 max:0.0684409 sd:0.0159295 dtype:DTYPE_FLOAT32 181: "text.180_of_512" NUMERICAL mean:0.0023879 min:-0.0859889 max:0.0903246 sd:0.040588 dtype:DTYPE_FLOAT32 182: "text.181_of_512" NUMERICAL mean:0.0469474 min:-0.0959887 max:0.0976843 sd:0.0347721 dtype:DTYPE_FLOAT32 183: "text.182_of_512" NUMERICAL mean:-0.00693341 min:-0.0887444 max:0.0929308 sd:0.0364697 dtype:DTYPE_FLOAT32 184: "text.183_of_512" NUMERICAL mean:-0.00740454 min:-0.0836568 max:0.0982687 sd:0.0353784 dtype:DTYPE_FLOAT32 185: "text.184_of_512" NUMERICAL mean:-0.0168595 min:-0.10092 max:0.0790249 sd:0.0388003 dtype:DTYPE_FLOAT32 186: "text.185_of_512" NUMERICAL mean:-0.00232853 min:-0.086526 max:0.0889761 sd:0.0353296 dtype:DTYPE_FLOAT32 187: "text.186_of_512" NUMERICAL mean:0.000582756 min:-0.0800429 max:0.0890964 sd:0.0342272 dtype:DTYPE_FLOAT32 188: "text.187_of_512" NUMERICAL mean:-0.00690847 min:-0.0788984 max:0.0999311 sd:0.0298525 dtype:DTYPE_FLOAT32 189: "text.188_of_512" NUMERICAL mean:0.067128 min:-0.073249 max:0.109486 sd:0.0164544 dtype:DTYPE_FLOAT32 190: "text.189_of_512" NUMERICAL mean:-0.0156762 min:-0.0897574 max:0.0937896 sd:0.0322483 dtype:DTYPE_FLOAT32 191: "text.190_of_512" NUMERICAL mean:0.020657 min:-0.0843347 max:0.0949148 sd:0.0366026 dtype:DTYPE_FLOAT32 192: "text.191_of_512" NUMERICAL mean:0.00570313 min:-0.0793632 max:0.0897408 sd:0.0385802 dtype:DTYPE_FLOAT32 193: "text.192_of_512" NUMERICAL mean:-0.0217683 min:-0.0857433 max:0.0830115 sd:0.0319114 dtype:DTYPE_FLOAT32 194: "text.193_of_512" NUMERICAL mean:-0.06861 min:-0.108961 max:0.0717299 sd:0.0141235 dtype:DTYPE_FLOAT32 195: "text.194_of_512" NUMERICAL mean:-0.0114472 min:-0.0918858 max:0.0872986 sd:0.0374338 dtype:DTYPE_FLOAT32 196: "text.195_of_512" NUMERICAL mean:-0.0242289 min:-0.0825904 max:0.0902977 sd:0.0309893 dtype:DTYPE_FLOAT32 197: "text.196_of_512" NUMERICAL mean:0.0261313 min:-0.0825517 max:0.0837058 sd:0.0314527 dtype:DTYPE_FLOAT32 198: "text.197_of_512" NUMERICAL mean:0.00897494 min:-0.0864797 max:0.0774307 sd:0.033463 dtype:DTYPE_FLOAT32 199: "text.198_of_512" NUMERICAL mean:0.0461119 min:-0.0768736 max:0.115535 sd:0.0328271 dtype:DTYPE_FLOAT32 200: "text.199_of_512" NUMERICAL mean:0.00238293 min:-0.0862734 max:0.115682 sd:0.0347315 dtype:DTYPE_FLOAT32 201: "text.200_of_512" NUMERICAL mean:-0.0217259 min:-0.0966809 max:0.0798758 sd:0.0401584 dtype:DTYPE_FLOAT32 202: "text.201_of_512" NUMERICAL mean:0.0206674 min:-0.0905074 max:0.0846099 sd:0.0348974 dtype:DTYPE_FLOAT32 203: "text.202_of_512" NUMERICAL mean:0.0106932 min:-0.0922981 max:0.0936093 sd:0.0362359 dtype:DTYPE_FLOAT32 204: "text.203_of_512" NUMERICAL mean:-0.000531681 min:-0.095827 max:0.113626 sd:0.0538437 dtype:DTYPE_FLOAT32 205: "text.204_of_512" NUMERICAL mean:0.0336588 min:-0.0854926 max:0.108485 sd:0.0316903 dtype:DTYPE_FLOAT32 206: "text.205_of_512" NUMERICAL mean:0.0026519 min:-0.0946099 max:0.0850502 sd:0.0352056 dtype:DTYPE_FLOAT32 207: "text.206_of_512" NUMERICAL mean:-0.00452261 min:-0.0915109 max:0.0820429 sd:0.0323585 dtype:DTYPE_FLOAT32 208: "text.207_of_512" NUMERICAL mean:-0.0281683 min:-0.0853697 max:0.0948644 sd:0.0371493 dtype:DTYPE_FLOAT32 209: "text.208_of_512" NUMERICAL mean:-0.0115137 min:-0.0833375 max:0.0836399 sd:0.0350201 dtype:DTYPE_FLOAT32 210: "text.209_of_512" NUMERICAL mean:0.0082079 min:-0.0866183 max:0.0800219 sd:0.0348183 dtype:DTYPE_FLOAT32 211: "text.210_of_512" NUMERICAL mean:-0.0377666 min:-0.0997254 max:0.0835936 sd:0.0305559 dtype:DTYPE_FLOAT32 212: "text.211_of_512" NUMERICAL mean:0.0143182 min:-0.0917944 max:0.0945138 sd:0.0352359 dtype:DTYPE_FLOAT32 213: "text.212_of_512" NUMERICAL mean:0.0196656 min:-0.0882236 max:0.0902131 sd:0.0358485 dtype:DTYPE_FLOAT32 214: "text.213_of_512" NUMERICAL mean:0.033878 min:-0.0713873 max:0.104101 sd:0.0283362 dtype:DTYPE_FLOAT32 215: "text.214_of_512" NUMERICAL mean:-0.0455993 min:-0.103323 max:0.0737987 sd:0.0271099 dtype:DTYPE_FLOAT32 216: "text.215_of_512" NUMERICAL mean:0.0126062 min:-0.0853854 max:0.0866744 sd:0.0348087 dtype:DTYPE_FLOAT32 217: "text.216_of_512" NUMERICAL mean:-0.0215749 min:-0.0906264 max:0.0845727 sd:0.0328653 dtype:DTYPE_FLOAT32 218: "text.217_of_512" NUMERICAL mean:-0.0165388 min:-0.106066 max:0.0853449 sd:0.0395682 dtype:DTYPE_FLOAT32 219: "text.218_of_512" NUMERICAL mean:0.0056953 min:-0.0889205 max:0.0880319 sd:0.0365522 dtype:DTYPE_FLOAT32 220: "text.219_of_512" NUMERICAL mean:-0.00380571 min:-0.0857684 max:0.0842541 sd:0.038037 dtype:DTYPE_FLOAT32 221: "text.220_of_512" NUMERICAL mean:0.00669877 min:-0.094601 max:0.0875136 sd:0.0359309 dtype:DTYPE_FLOAT32 222: "text.221_of_512" NUMERICAL mean:-0.00228978 min:-0.0838995 max:0.0871175 sd:0.0380861 dtype:DTYPE_FLOAT32 223: "text.222_of_512" NUMERICAL mean:-0.0148104 min:-0.0843962 max:0.0797233 sd:0.0328315 dtype:DTYPE_FLOAT32 224: "text.223_of_512" NUMERICAL mean:-0.00308675 min:-0.0867925 max:0.0861399 sd:0.0367905 dtype:DTYPE_FLOAT32 225: "text.224_of_512" NUMERICAL mean:-0.0651968 min:-0.112212 max:0.108484 sd:0.025093 dtype:DTYPE_FLOAT32 226: "text.225_of_512" NUMERICAL mean:-0.0355706 min:-0.0940091 max:0.0825248 sd:0.0294568 dtype:DTYPE_FLOAT32 227: "text.226_of_512" NUMERICAL mean:-0.0371873 min:-0.111241 max:0.0823205 sd:0.0404316 dtype:DTYPE_FLOAT32 228: "text.227_of_512" NUMERICAL mean:-0.00956319 min:-0.0947813 max:0.0932219 sd:0.0380526 dtype:DTYPE_FLOAT32 229: "text.228_of_512" NUMERICAL mean:0.00753233 min:-0.0797382 max:0.0850486 sd:0.0362034 dtype:DTYPE_FLOAT32 230: "text.229_of_512" NUMERICAL mean:-0.0411396 min:-0.093706 max:0.0783683 sd:0.0275233 dtype:DTYPE_FLOAT32 231: "text.230_of_512" NUMERICAL mean:0.0257311 min:-0.0813399 max:0.0931238 sd:0.0344775 dtype:DTYPE_FLOAT32 232: "text.231_of_512" NUMERICAL mean:0.00856159 min:-0.0874029 max:0.0896917 sd:0.0328607 dtype:DTYPE_FLOAT32 233: "text.232_of_512" NUMERICAL mean:0.00799258 min:-0.0823268 max:0.0926399 sd:0.0355411 dtype:DTYPE_FLOAT32 234: "text.233_of_512" NUMERICAL mean:-0.0760438 min:-0.111486 max:0.0816572 sd:0.0086788 dtype:DTYPE_FLOAT32 235: "text.234_of_512" NUMERICAL mean:0.0105768 min:-0.0819705 max:0.077538 sd:0.0310803 dtype:DTYPE_FLOAT32 236: "text.235_of_512" NUMERICAL mean:-0.00394976 min:-0.081968 max:0.0948637 sd:0.0342805 dtype:DTYPE_FLOAT32 237: "text.236_of_512" NUMERICAL mean:-0.0060946 min:-0.0855534 max:0.0784498 sd:0.0302966 dtype:DTYPE_FLOAT32 238: "text.237_of_512" NUMERICAL mean:-0.00495043 min:-0.0832484 max:0.0864264 sd:0.0371133 dtype:DTYPE_FLOAT32 239: "text.238_of_512" NUMERICAL mean:-0.0355637 min:-0.104853 max:0.0651103 sd:0.0246829 dtype:DTYPE_FLOAT32 240: "text.239_of_512" NUMERICAL mean:0.0270039 min:-0.0820892 max:0.0987316 sd:0.0326327 dtype:DTYPE_FLOAT32 241: "text.240_of_512" NUMERICAL mean:0.0288838 min:-0.0861977 max:0.0915473 sd:0.0318938 dtype:DTYPE_FLOAT32 242: "text.241_of_512" NUMERICAL mean:-0.00962749 min:-0.0856368 max:0.0814245 sd:0.0344315 dtype:DTYPE_FLOAT32 243: "text.242_of_512" NUMERICAL mean:0.00896494 min:-0.0951616 max:0.0850005 sd:0.0413986 dtype:DTYPE_FLOAT32 244: "text.243_of_512" NUMERICAL mean:-0.0226355 min:-0.0881037 max:0.0938583 sd:0.0385061 dtype:DTYPE_FLOAT32 245: "text.244_of_512" NUMERICAL mean:0.0241801 min:-0.0883454 max:0.0911317 sd:0.0348305 dtype:DTYPE_FLOAT32 246: "text.245_of_512" NUMERICAL mean:0.0254446 min:-0.0901449 max:0.107281 sd:0.039955 dtype:DTYPE_FLOAT32 247: "text.246_of_512" NUMERICAL mean:-0.0213719 min:-0.0852335 max:0.082019 sd:0.0325582 dtype:DTYPE_FLOAT32 248: "text.247_of_512" NUMERICAL mean:-0.00991342 min:-0.0842142 max:0.0863353 sd:0.0365995 dtype:DTYPE_FLOAT32 249: "text.248_of_512" NUMERICAL mean:0.00749679 min:-0.112647 max:0.100322 sd:0.0392081 dtype:DTYPE_FLOAT32 250: "text.249_of_512" NUMERICAL mean:-0.00762937 min:-0.0943137 max:0.0774691 sd:0.0362134 dtype:DTYPE_FLOAT32 251: "text.250_of_512" NUMERICAL mean:0.0389247 min:-0.0973366 max:0.0903543 sd:0.0333933 dtype:DTYPE_FLOAT32 252: "text.251_of_512" NUMERICAL mean:-0.044424 min:-0.0998835 max:0.0824921 sd:0.0288488 dtype:DTYPE_FLOAT32 253: "text.252_of_512" NUMERICAL mean:0.0251871 min:-0.084684 max:0.0898936 sd:0.0347433 dtype:DTYPE_FLOAT32 254: "text.253_of_512" NUMERICAL mean:-0.0189432 min:-0.0888732 max:0.0819862 sd:0.0337242 dtype:DTYPE_FLOAT32 255: "text.254_of_512" NUMERICAL mean:0.0350538 min:-0.085493 max:0.100748 sd:0.0344377 dtype:DTYPE_FLOAT32 256: "text.255_of_512" NUMERICAL mean:0.0153939 min:-0.0889486 max:0.090479 sd:0.0361216 dtype:DTYPE_FLOAT32 257: "text.256_of_512" NUMERICAL mean:0.0118623 min:-0.0846893 max:0.0877129 sd:0.036393 dtype:DTYPE_FLOAT32 258: "text.257_of_512" NUMERICAL mean:0.07627 min:-0.00150217 max:0.121376 sd:0.0086155 dtype:DTYPE_FLOAT32 259: "text.258_of_512" NUMERICAL mean:0.0379985 min:-0.0748384 max:0.0967968 sd:0.0292381 dtype:DTYPE_FLOAT32 260: "text.259_of_512" NUMERICAL mean:0.0159811 min:-0.0787165 max:0.0932967 sd:0.0354853 dtype:DTYPE_FLOAT32 261: "text.260_of_512" NUMERICAL mean:0.00576277 min:-0.0836212 max:0.0860075 sd:0.0310029 dtype:DTYPE_FLOAT32 262: "text.261_of_512" NUMERICAL mean:0.000585571 min:-0.0844256 max:0.0841413 sd:0.0345458 dtype:DTYPE_FLOAT32 263: "text.262_of_512" NUMERICAL mean:-0.00469284 min:-0.088471 max:0.110526 sd:0.0379816 dtype:DTYPE_FLOAT32 264: "text.263_of_512" NUMERICAL mean:-0.0144205 min:-0.0941174 max:0.0727047 sd:0.0317815 dtype:DTYPE_FLOAT32 265: "text.264_of_512" NUMERICAL mean:-0.00347019 min:-0.0876852 max:0.0943233 sd:0.0364456 dtype:DTYPE_FLOAT32 266: "text.265_of_512" NUMERICAL mean:0.0179998 min:-0.0831757 max:0.0890779 sd:0.0336362 dtype:DTYPE_FLOAT32 267: "text.266_of_512" NUMERICAL mean:-0.00646691 min:-0.0895957 max:0.0862346 sd:0.0353584 dtype:DTYPE_FLOAT32 268: "text.267_of_512" NUMERICAL mean:0.0294141 min:-0.0832922 max:0.095093 sd:0.0347372 dtype:DTYPE_FLOAT32 269: "text.268_of_512" NUMERICAL mean:0.01249 min:-0.0856392 max:0.0893549 sd:0.0367725 dtype:DTYPE_FLOAT32 270: "text.269_of_512" NUMERICAL mean:-0.0162131 min:-0.0933031 max:0.0840509 sd:0.0361621 dtype:DTYPE_FLOAT32 271: "text.270_of_512" NUMERICAL mean:-0.012055 min:-0.0927456 max:0.0888101 sd:0.0339707 dtype:DTYPE_FLOAT32 272: "text.271_of_512" NUMERICAL mean:0.0616041 min:-0.0714294 max:0.107679 sd:0.0185813 dtype:DTYPE_FLOAT32 273: "text.272_of_512" NUMERICAL mean:-0.00189753 min:-0.0896988 max:0.0844877 sd:0.0361814 dtype:DTYPE_FLOAT32 274: "text.273_of_512" NUMERICAL mean:-0.0112785 min:-0.0849119 max:0.0824402 sd:0.036787 dtype:DTYPE_FLOAT32 275: "text.274_of_512" NUMERICAL mean:-0.0215319 min:-0.0950761 max:0.0831872 sd:0.0347278 dtype:DTYPE_FLOAT32 276: "text.275_of_512" NUMERICAL mean:-0.0079228 min:-0.0799229 max:0.0998652 sd:0.0360374 dtype:DTYPE_FLOAT32 277: "text.276_of_512" NUMERICAL mean:-0.0303123 min:-0.0916027 max:0.0841094 sd:0.0360064 dtype:DTYPE_FLOAT32 278: "text.277_of_512" NUMERICAL mean:-0.0492444 min:-0.107444 max:0.0758573 sd:0.0270125 dtype:DTYPE_FLOAT32 279: "text.278_of_512" NUMERICAL mean:0.0655957 min:-0.070574 max:0.11492 sd:0.0186785 dtype:DTYPE_FLOAT32 280: "text.279_of_512" NUMERICAL mean:0.0450869 min:-0.0835811 max:0.104559 sd:0.0343092 dtype:DTYPE_FLOAT32 281: "text.280_of_512" NUMERICAL mean:-0.00275616 min:-0.0926189 max:0.0863733 sd:0.0417567 dtype:DTYPE_FLOAT32 282: "text.281_of_512" NUMERICAL mean:-0.0225917 min:-0.0885091 max:0.0831179 sd:0.0349756 dtype:DTYPE_FLOAT32 283: "text.282_of_512" NUMERICAL mean:-0.0135245 min:-0.0841457 max:0.091283 sd:0.0333958 dtype:DTYPE_FLOAT32 284: "text.283_of_512" NUMERICAL mean:0.0120968 min:-0.0832972 max:0.0963631 sd:0.0372584 dtype:DTYPE_FLOAT32 285: "text.284_of_512" NUMERICAL mean:-0.00136727 min:-0.0837919 max:0.0915016 sd:0.0385175 dtype:DTYPE_FLOAT32 286: "text.285_of_512" NUMERICAL mean:-0.0745365 min:-0.11774 max:0.0824989 sd:0.0131683 dtype:DTYPE_FLOAT32 287: "text.286_of_512" NUMERICAL mean:0.00845783 min:-0.089365 max:0.0842921 sd:0.0366287 dtype:DTYPE_FLOAT32 288: "text.287_of_512" NUMERICAL mean:0.0176049 min:-0.0813912 max:0.0945282 sd:0.0349949 dtype:DTYPE_FLOAT32 289: "text.288_of_512" NUMERICAL mean:-0.00175854 min:-0.0919866 max:0.090152 sd:0.0382503 dtype:DTYPE_FLOAT32 290: "text.289_of_512" NUMERICAL mean:0.0414527 min:-0.0834453 max:0.0964402 sd:0.0279968 dtype:DTYPE_FLOAT32 291: "text.290_of_512" NUMERICAL mean:-0.0237126 min:-0.0890423 max:0.0865852 sd:0.0367122 dtype:DTYPE_FLOAT32 292: "text.291_of_512" NUMERICAL mean:0.0706647 min:-0.0659814 max:0.111782 sd:0.0142339 dtype:DTYPE_FLOAT32 293: "text.292_of_512" NUMERICAL mean:0.0428567 min:-0.0927111 max:0.088533 sd:0.0327827 dtype:DTYPE_FLOAT32 294: "text.293_of_512" NUMERICAL mean:0.0128566 min:-0.086611 max:0.0913341 sd:0.0368175 dtype:DTYPE_FLOAT32 295: "text.294_of_512" NUMERICAL mean:0.0405849 min:-0.0812254 max:0.111267 sd:0.0298301 dtype:DTYPE_FLOAT32 296: "text.295_of_512" NUMERICAL mean:0.00254073 min:-0.0844188 max:0.0850672 sd:0.0312665 dtype:DTYPE_FLOAT32 297: "text.296_of_512" NUMERICAL mean:-0.000922522 min:-0.108404 max:0.09455 sd:0.0461599 dtype:DTYPE_FLOAT32 298: "text.297_of_512" NUMERICAL mean:-0.00948435 min:-0.0960434 max:0.0809376 sd:0.0354375 dtype:DTYPE_FLOAT32 299: "text.298_of_512" NUMERICAL mean:-0.00522944 min:-0.0953566 max:0.0807673 sd:0.038705 dtype:DTYPE_FLOAT32 300: "text.299_of_512" NUMERICAL mean:0.0469335 min:-0.0978255 max:0.0942754 sd:0.0349978 dtype:DTYPE_FLOAT32 301: "text.300_of_512" NUMERICAL mean:0.0136288 min:-0.0887664 max:0.088646 sd:0.0358246 dtype:DTYPE_FLOAT32 302: "text.301_of_512" NUMERICAL mean:-0.00768847 min:-0.0876874 max:0.0863244 sd:0.0365417 dtype:DTYPE_FLOAT32 303: "text.302_of_512" NUMERICAL mean:0.0255685 min:-0.0882509 max:0.0889251 sd:0.0293385 dtype:DTYPE_FLOAT32 304: "text.303_of_512" NUMERICAL mean:0.00278844 min:-0.0815436 max:0.0932456 sd:0.0356314 dtype:DTYPE_FLOAT32 305: "text.304_of_512" NUMERICAL mean:0.0189678 min:-0.0837086 max:0.0896033 sd:0.03269 dtype:DTYPE_FLOAT32 306: "text.305_of_512" NUMERICAL mean:-0.0190712 min:-0.0865702 max:0.0826589 sd:0.0336245 dtype:DTYPE_FLOAT32 307: "text.306_of_512" NUMERICAL mean:0.00711986 min:-0.0865388 max:0.0890463 sd:0.0373572 dtype:DTYPE_FLOAT32 308: "text.307_of_512" NUMERICAL mean:-0.0249224 min:-0.0874085 max:0.0792676 sd:0.0325445 dtype:DTYPE_FLOAT32 309: "text.308_of_512" NUMERICAL mean:0.0283369 min:-0.0824256 max:0.093054 sd:0.0325861 dtype:DTYPE_FLOAT32 310: "text.309_of_512" NUMERICAL mean:0.0220595 min:-0.0836122 max:0.0915141 sd:0.0359587 dtype:DTYPE_FLOAT32 311: "text.310_of_512" NUMERICAL mean:0.0117776 min:-0.0932258 max:0.0915613 sd:0.0362803 dtype:DTYPE_FLOAT32 312: "text.311_of_512" NUMERICAL mean:-0.0221651 min:-0.0819381 max:0.0840218 sd:0.0312305 dtype:DTYPE_FLOAT32 313: "text.312_of_512" NUMERICAL mean:0.00199115 min:-0.099875 max:0.0830933 sd:0.0360407 dtype:DTYPE_FLOAT32 314: "text.313_of_512" NUMERICAL mean:0.00353999 min:-0.085414 max:0.085891 sd:0.0352858 dtype:DTYPE_FLOAT32 315: "text.314_of_512" NUMERICAL mean:0.0105751 min:-0.0887366 max:0.0830806 sd:0.0344889 dtype:DTYPE_FLOAT32 316: "text.315_of_512" NUMERICAL mean:-0.00350684 min:-0.0834666 max:0.0800134 sd:0.0353813 dtype:DTYPE_FLOAT32 317: "text.316_of_512" NUMERICAL mean:0.00208485 min:-0.0947385 max:0.0856062 sd:0.037774 dtype:DTYPE_FLOAT32 318: "text.317_of_512" NUMERICAL mean:0.0249101 min:-0.0906462 max:0.0893748 sd:0.0373777 dtype:DTYPE_FLOAT32 319: "text.318_of_512" NUMERICAL mean:0.0493711 min:-0.0869199 max:0.114657 sd:0.0328398 dtype:DTYPE_FLOAT32 320: "text.319_of_512" NUMERICAL mean:-0.00762606 min:-0.0862671 max:0.0973113 sd:0.0373719 dtype:DTYPE_FLOAT32 321: "text.320_of_512" NUMERICAL mean:-0.0253199 min:-0.0864342 max:0.102423 sd:0.035196 dtype:DTYPE_FLOAT32 322: "text.321_of_512" NUMERICAL mean:-0.00615397 min:-0.0893151 max:0.0887937 sd:0.0395088 dtype:DTYPE_FLOAT32 323: "text.322_of_512" NUMERICAL mean:-0.0198416 min:-0.0808634 max:0.0972422 sd:0.030688 dtype:DTYPE_FLOAT32 324: "text.323_of_512" NUMERICAL mean:-0.0335036 min:-0.0842939 max:0.0986979 sd:0.0349683 dtype:DTYPE_FLOAT32 325: "text.324_of_512" NUMERICAL mean:-0.00820786 min:-0.0859345 max:0.090916 sd:0.0353986 dtype:DTYPE_FLOAT32 326: "text.325_of_512" NUMERICAL mean:0.0100534 min:-0.0919951 max:0.0962937 sd:0.0443805 dtype:DTYPE_FLOAT32 327: "text.326_of_512" NUMERICAL mean:-0.00313179 min:-0.0874311 max:0.0892223 sd:0.0355988 dtype:DTYPE_FLOAT32 328: "text.327_of_512" NUMERICAL mean:0.0213434 min:-0.0810096 max:0.0901993 sd:0.0378811 dtype:DTYPE_FLOAT32 329: "text.328_of_512" NUMERICAL mean:-0.0250477 min:-0.0960049 max:0.0842342 sd:0.0314437 dtype:DTYPE_FLOAT32 330: "text.329_of_512" NUMERICAL mean:0.00248319 min:-0.0942121 max:0.0882779 sd:0.0373737 dtype:DTYPE_FLOAT32 331: "text.330_of_512" NUMERICAL mean:0.0101506 min:-0.09232 max:0.0812322 sd:0.0344001 dtype:DTYPE_FLOAT32 332: "text.331_of_512" NUMERICAL mean:0.00491259 min:-0.0886715 max:0.0877843 sd:0.037935 dtype:DTYPE_FLOAT32 333: "text.332_of_512" NUMERICAL mean:0.0122141 min:-0.088813 max:0.0847952 sd:0.0371118 dtype:DTYPE_FLOAT32 334: "text.333_of_512" NUMERICAL mean:0.0241871 min:-0.102588 max:0.094858 sd:0.0306593 dtype:DTYPE_FLOAT32 335: "text.334_of_512" NUMERICAL mean:-0.0632664 min:-0.109776 max:0.0803932 sd:0.0211622 dtype:DTYPE_FLOAT32 336: "text.335_of_512" NUMERICAL mean:-0.013856 min:-0.08953 max:0.0885321 sd:0.0327168 dtype:DTYPE_FLOAT32 337: "text.336_of_512" NUMERICAL mean:-0.0136259 min:-0.0984544 max:0.0779921 sd:0.0357739 dtype:DTYPE_FLOAT32 338: "text.337_of_512" NUMERICAL mean:-0.0106961 min:-0.089448 max:0.0899969 sd:0.0376812 dtype:DTYPE_FLOAT32 339: "text.338_of_512" NUMERICAL mean:0.0199777 min:-0.0840479 max:0.111438 sd:0.042303 dtype:DTYPE_FLOAT32 340: "text.339_of_512" NUMERICAL mean:-0.0218739 min:-0.0883652 max:0.079979 sd:0.0347288 dtype:DTYPE_FLOAT32 341: "text.340_of_512" NUMERICAL mean:0.06935 min:-0.0726485 max:0.115548 sd:0.0155241 dtype:DTYPE_FLOAT32 342: "text.341_of_512" NUMERICAL mean:0.0310573 min:-0.0919902 max:0.0881952 sd:0.0364025 dtype:DTYPE_FLOAT32 343: "text.342_of_512" NUMERICAL mean:0.000973466 min:-0.0857906 max:0.0862984 sd:0.0384532 dtype:DTYPE_FLOAT32 344: "text.343_of_512" NUMERICAL mean:0.00264308 min:-0.0844924 max:0.085991 sd:0.0363975 dtype:DTYPE_FLOAT32 345: "text.344_of_512" NUMERICAL mean:-0.00907527 min:-0.079237 max:0.0723712 sd:0.0283499 dtype:DTYPE_FLOAT32 346: "text.345_of_512" NUMERICAL mean:-0.0228477 min:-0.0984037 max:0.0765738 sd:0.0344783 dtype:DTYPE_FLOAT32 347: "text.346_of_512" NUMERICAL mean:-0.00742428 min:-0.0987589 max:0.0825624 sd:0.036547 dtype:DTYPE_FLOAT32 348: "text.347_of_512" NUMERICAL mean:-0.0586682 min:-0.115158 max:0.0828087 sd:0.0299437 dtype:DTYPE_FLOAT32 349: "text.348_of_512" NUMERICAL mean:-0.00689071 min:-0.0987143 max:0.0862847 sd:0.0359991 dtype:DTYPE_FLOAT32 350: "text.349_of_512" NUMERICAL mean:0.0348556 min:-0.0816546 max:0.0964646 sd:0.0322193 dtype:DTYPE_FLOAT32 351: "text.350_of_512" NUMERICAL mean:-0.00442726 min:-0.0845818 max:0.0790038 sd:0.035079 dtype:DTYPE_FLOAT32 352: "text.351_of_512" NUMERICAL mean:0.000284563 min:-0.0960373 max:0.0851551 sd:0.0376517 dtype:DTYPE_FLOAT32 353: "text.352_of_512" NUMERICAL mean:-0.00134722 min:-0.0890189 max:0.0820552 sd:0.0348834 dtype:DTYPE_FLOAT32 354: "text.353_of_512" NUMERICAL mean:-0.0121378 min:-0.0891658 max:0.088453 sd:0.0355917 dtype:DTYPE_FLOAT32 355: "text.354_of_512" NUMERICAL mean:-0.0188861 min:-0.109144 max:0.0795723 sd:0.0361681 dtype:DTYPE_FLOAT32 356: "text.355_of_512" NUMERICAL mean:-0.0491884 min:-0.10172 max:0.0680702 sd:0.0255385 dtype:DTYPE_FLOAT32 357: "text.356_of_512" NUMERICAL mean:-0.0758847 min:-0.120116 max:-0.0497331 sd:0.00807918 dtype:DTYPE_FLOAT32 358: "text.357_of_512" NUMERICAL mean:-0.00333743 min:-0.0904124 max:0.0876658 sd:0.037835 dtype:DTYPE_FLOAT32 359: "text.358_of_512" NUMERICAL mean:-0.00367021 min:-0.0876946 max:0.0955447 sd:0.037389 dtype:DTYPE_FLOAT32 360: "text.359_of_512" NUMERICAL mean:-0.00055956 min:-0.0904881 max:0.0875923 sd:0.0388563 dtype:DTYPE_FLOAT32 361: "text.360_of_512" NUMERICAL mean:0.00259386 min:-0.108261 max:0.0841461 sd:0.0381267 dtype:DTYPE_FLOAT32 362: "text.361_of_512" NUMERICAL mean:-0.0241437 min:-0.0890906 max:0.0928297 sd:0.0366722 dtype:DTYPE_FLOAT32 363: "text.362_of_512" NUMERICAL mean:0.0739337 min:-0.0656038 max:0.128404 sd:0.010182 dtype:DTYPE_FLOAT32 364: "text.363_of_512" NUMERICAL mean:0.0508201 min:-0.0864671 max:0.083986 sd:0.0211588 dtype:DTYPE_FLOAT32 365: "text.364_of_512" NUMERICAL mean:0.0226045 min:-0.10529 max:0.0848131 sd:0.0387468 dtype:DTYPE_FLOAT32 366: "text.365_of_512" NUMERICAL mean:-0.0256886 min:-0.100561 max:0.0940806 sd:0.0342923 dtype:DTYPE_FLOAT32 367: "text.366_of_512" NUMERICAL mean:0.0655513 min:-0.0860083 max:0.103827 sd:0.0171798 dtype:DTYPE_FLOAT32 368: "text.367_of_512" NUMERICAL mean:0.000812781 min:-0.0851833 max:0.0878781 sd:0.0357148 dtype:DTYPE_FLOAT32 369: "text.368_of_512" NUMERICAL mean:0.0347878 min:-0.0734774 max:0.105422 sd:0.0276884 dtype:DTYPE_FLOAT32 370: "text.369_of_512" NUMERICAL mean:-0.0097847 min:-0.0879691 max:0.0830986 sd:0.0357574 dtype:DTYPE_FLOAT32 371: "text.370_of_512" NUMERICAL mean:-0.00668312 min:-0.0830364 max:0.0887912 sd:0.0368456 dtype:DTYPE_FLOAT32 372: "text.371_of_512" NUMERICAL mean:0.0129623 min:-0.0856702 max:0.0906184 sd:0.0355013 dtype:DTYPE_FLOAT32 373: "text.372_of_512" NUMERICAL mean:-0.0293164 min:-0.0912901 max:0.0773753 sd:0.0324155 dtype:DTYPE_FLOAT32 374: "text.373_of_512" NUMERICAL mean:0.0474531 min:-0.0852198 max:0.0912352 sd:0.0275962 dtype:DTYPE_FLOAT32 375: "text.374_of_512" NUMERICAL mean:-0.0163036 min:-0.0825631 max:0.0919325 sd:0.0344408 dtype:DTYPE_FLOAT32 376: "text.375_of_512" NUMERICAL mean:0.034497 min:-0.0848398 max:0.111058 sd:0.0346333 dtype:DTYPE_FLOAT32 377: "text.376_of_512" NUMERICAL mean:-0.00185094 min:-0.081474 max:0.0848511 sd:0.0368544 dtype:DTYPE_FLOAT32 378: "text.377_of_512" NUMERICAL mean:-0.0127358 min:-0.0948508 max:0.0787196 sd:0.0360913 dtype:DTYPE_FLOAT32 379: "text.378_of_512" NUMERICAL mean:-0.00578751 min:-0.0924906 max:0.0861703 sd:0.035065 dtype:DTYPE_FLOAT32 380: "text.379_of_512" NUMERICAL mean:-0.0471052 min:-0.0949595 max:0.0911769 sd:0.0322388 dtype:DTYPE_FLOAT32 381: "text.380_of_512" NUMERICAL mean:-0.00668082 min:-0.0873622 max:0.0909038 sd:0.0390076 dtype:DTYPE_FLOAT32 382: "text.381_of_512" NUMERICAL mean:-0.00774684 min:-0.091025 max:0.0877729 sd:0.0351947 dtype:DTYPE_FLOAT32 383: "text.382_of_512" NUMERICAL mean:0.0286156 min:-0.0806168 max:0.0918561 sd:0.0332708 dtype:DTYPE_FLOAT32 384: "text.383_of_512" NUMERICAL mean:-0.0110346 min:-0.0946519 max:0.0829471 sd:0.0352531 dtype:DTYPE_FLOAT32 385: "text.384_of_512" NUMERICAL mean:0.0305969 min:-0.0852557 max:0.101585 sd:0.0334592 dtype:DTYPE_FLOAT32 386: "text.385_of_512" NUMERICAL mean:-0.0713245 min:-0.107546 max:0.0513883 sd:0.0102852 dtype:DTYPE_FLOAT32 387: "text.386_of_512" NUMERICAL mean:0.0383743 min:-0.0951244 max:0.0995259 sd:0.0384113 dtype:DTYPE_FLOAT32 388: "text.387_of_512" NUMERICAL mean:0.0214273 min:-0.082834 max:0.0837052 sd:0.0331753 dtype:DTYPE_FLOAT32 389: "text.388_of_512" NUMERICAL mean:0.0190755 min:-0.0825616 max:0.0880756 sd:0.0358801 dtype:DTYPE_FLOAT32 390: "text.389_of_512" NUMERICAL mean:-0.0121952 min:-0.0867737 max:0.0962556 sd:0.0377747 dtype:DTYPE_FLOAT32 391: "text.390_of_512" NUMERICAL mean:0.0437497 min:-0.10375 max:0.104252 sd:0.0328493 dtype:DTYPE_FLOAT32 392: "text.391_of_512" NUMERICAL mean:0.00412824 min:-0.0841954 max:0.093423 sd:0.0395505 dtype:DTYPE_FLOAT32 393: "text.392_of_512" NUMERICAL mean:0.0346048 min:-0.0794608 max:0.0912714 sd:0.0322768 dtype:DTYPE_FLOAT32 394: "text.393_of_512" NUMERICAL mean:-0.00179405 min:-0.0847392 max:0.084127 sd:0.0355668 dtype:DTYPE_FLOAT32 395: "text.394_of_512" NUMERICAL mean:0.0111267 min:-0.0895413 max:0.0899808 sd:0.0379719 dtype:DTYPE_FLOAT32 396: "text.395_of_512" NUMERICAL mean:-0.00366774 min:-0.0787723 max:0.0990571 sd:0.0380017 dtype:DTYPE_FLOAT32 397: "text.396_of_512" NUMERICAL mean:0.011426 min:-0.0846937 max:0.115535 sd:0.0497252 dtype:DTYPE_FLOAT32 398: "text.397_of_512" NUMERICAL mean:-0.000477306 min:-0.0886949 max:0.0863151 sd:0.0385991 dtype:DTYPE_FLOAT32 399: "text.398_of_512" NUMERICAL mean:-0.0180848 min:-0.115534 max:0.0796439 sd:0.0413374 dtype:DTYPE_FLOAT32 400: "text.399_of_512" NUMERICAL mean:-0.0595741 min:-0.114342 max:0.0863008 sd:0.0276777 dtype:DTYPE_FLOAT32 401: "text.400_of_512" NUMERICAL mean:0.0257502 min:-0.0852726 max:0.0849641 sd:0.0320993 dtype:DTYPE_FLOAT32 402: "text.401_of_512" NUMERICAL mean:-0.0119989 min:-0.0848179 max:0.0853326 sd:0.0342696 dtype:DTYPE_FLOAT32 403: "text.402_of_512" NUMERICAL mean:0.00456497 min:-0.0891429 max:0.090027 sd:0.0356174 dtype:DTYPE_FLOAT32 404: "text.403_of_512" NUMERICAL mean:-0.0278144 min:-0.0943406 max:0.0744098 sd:0.0318229 dtype:DTYPE_FLOAT32 405: "text.404_of_512" NUMERICAL mean:0.0742517 min:-0.0737848 max:0.120351 sd:0.013182 dtype:DTYPE_FLOAT32 406: "text.405_of_512" NUMERICAL mean:-0.00356677 min:-0.083284 max:0.085719 sd:0.0373844 dtype:DTYPE_FLOAT32 407: "text.406_of_512" NUMERICAL mean:0.0732326 min:-0.0643611 max:0.116212 sd:0.0124106 dtype:DTYPE_FLOAT32 408: "text.407_of_512" NUMERICAL mean:0.00309889 min:-0.0876725 max:0.0937136 sd:0.0342554 dtype:DTYPE_FLOAT32 409: "text.408_of_512" NUMERICAL mean:0.000529527 min:-0.108319 max:0.0765514 sd:0.0442172 dtype:DTYPE_FLOAT32 410: "text.409_of_512" NUMERICAL mean:-0.0265223 min:-0.0970284 max:0.0848186 sd:0.0365646 dtype:DTYPE_FLOAT32 411: "text.410_of_512" NUMERICAL mean:-0.0063123 min:-0.0907225 max:0.0766137 sd:0.0353277 dtype:DTYPE_FLOAT32 412: "text.411_of_512" NUMERICAL mean:0.0128983 min:-0.0890376 max:0.0832104 sd:0.0345936 dtype:DTYPE_FLOAT32 413: "text.412_of_512" NUMERICAL mean:0.0020441 min:-0.080098 max:0.0911496 sd:0.0373864 dtype:DTYPE_FLOAT32 414: "text.413_of_512" NUMERICAL mean:0.00372714 min:-0.0915223 max:0.0912589 sd:0.0363697 dtype:DTYPE_FLOAT32 415: "text.414_of_512" NUMERICAL mean:0.0591218 min:-0.0867741 max:0.114746 sd:0.0273598 dtype:DTYPE_FLOAT32 416: "text.415_of_512" NUMERICAL mean:-0.00308806 min:-0.0834403 max:0.0937164 sd:0.0396478 dtype:DTYPE_FLOAT32 417: "text.416_of_512" NUMERICAL mean:-0.0236163 min:-0.0869295 max:0.0845337 sd:0.0354895 dtype:DTYPE_FLOAT32 418: "text.417_of_512" NUMERICAL mean:-0.0135025 min:-0.0844859 max:0.0827825 sd:0.0336036 dtype:DTYPE_FLOAT32 419: "text.418_of_512" NUMERICAL mean:0.00223071 min:-0.0927571 max:0.0924417 sd:0.0382593 dtype:DTYPE_FLOAT32 420: "text.419_of_512" NUMERICAL mean:-0.0187637 min:-0.0875421 max:0.0888282 sd:0.0389555 dtype:DTYPE_FLOAT32 421: "text.420_of_512" NUMERICAL mean:0.0348483 min:-0.0838916 max:0.0912609 sd:0.0329786 dtype:DTYPE_FLOAT32 422: "text.421_of_512" NUMERICAL mean:0.00654405 min:-0.0999311 max:0.0859796 sd:0.0360056 dtype:DTYPE_FLOAT32 423: "text.422_of_512" NUMERICAL mean:-0.0413183 min:-0.103786 max:0.0723974 sd:0.0326804 dtype:DTYPE_FLOAT32 424: "text.423_of_512" NUMERICAL mean:-0.00331647 min:-0.0950863 max:0.0779905 sd:0.0349987 dtype:DTYPE_FLOAT32 425: "text.424_of_512" NUMERICAL mean:-0.00119706 min:-0.0864089 max:0.0919001 sd:0.0406214 dtype:DTYPE_FLOAT32 426: "text.425_of_512" NUMERICAL mean:0.0492668 min:-0.0742577 max:0.0987621 sd:0.0249189 dtype:DTYPE_FLOAT32 427: "text.426_of_512" NUMERICAL mean:-0.00389623 min:-0.0906745 max:0.0850097 sd:0.0378908 dtype:DTYPE_FLOAT32 428: "text.427_of_512" NUMERICAL mean:0.0661491 min:-0.0886513 max:0.0989667 sd:0.0167134 dtype:DTYPE_FLOAT32 429: "text.428_of_512" NUMERICAL mean:-0.0110098 min:-0.0842938 max:0.0861742 sd:0.0371607 dtype:DTYPE_FLOAT32 430: "text.429_of_512" NUMERICAL mean:0.0578028 min:-0.0859271 max:0.113576 sd:0.0302465 dtype:DTYPE_FLOAT32 431: "text.430_of_512" NUMERICAL mean:0.00414731 min:-0.0857654 max:0.084379 sd:0.0376122 dtype:DTYPE_FLOAT32 432: "text.431_of_512" NUMERICAL mean:0.0077961 min:-0.0895345 max:0.0873379 sd:0.0353009 dtype:DTYPE_FLOAT32 433: "text.432_of_512" NUMERICAL mean:-0.00979928 min:-0.0915835 max:0.0957983 sd:0.0376023 dtype:DTYPE_FLOAT32 434: "text.433_of_512" NUMERICAL mean:0.0558396 min:-0.0769163 max:0.11158 sd:0.0257219 dtype:DTYPE_FLOAT32 435: "text.434_of_512" NUMERICAL mean:-0.029334 min:-0.0929427 max:0.0768163 sd:0.0353737 dtype:DTYPE_FLOAT32 436: "text.435_of_512" NUMERICAL mean:-0.0258972 min:-0.107938 max:0.0752108 sd:0.0291947 dtype:DTYPE_FLOAT32 437: "text.436_of_512" NUMERICAL mean:0.0138365 min:-0.0798527 max:0.0904732 sd:0.0283121 dtype:DTYPE_FLOAT32 438: "text.437_of_512" NUMERICAL mean:0.00229088 min:-0.0916949 max:0.0918499 sd:0.0368933 dtype:DTYPE_FLOAT32 439: "text.438_of_512" NUMERICAL mean:-0.0150511 min:-0.0888081 max:0.0842706 sd:0.0346245 dtype:DTYPE_FLOAT32 440: "text.439_of_512" NUMERICAL mean:0.00617906 min:-0.0931821 max:0.081212 sd:0.0383849 dtype:DTYPE_FLOAT32 441: "text.440_of_512" NUMERICAL mean:0.0118221 min:-0.0908024 max:0.082485 sd:0.038097 dtype:DTYPE_FLOAT32 442: "text.441_of_512" NUMERICAL mean:-0.0457463 min:-0.11794 max:0.0860535 sd:0.0346978 dtype:DTYPE_FLOAT32 443: "text.442_of_512" NUMERICAL mean:0.035684 min:-0.0763339 max:0.0894163 sd:0.0270372 dtype:DTYPE_FLOAT32 444: "text.443_of_512" NUMERICAL mean:-0.00365348 min:-0.087954 max:0.0892831 sd:0.032805 dtype:DTYPE_FLOAT32 445: "text.444_of_512" NUMERICAL mean:0.00353574 min:-0.0873545 max:0.0859059 sd:0.0392823 dtype:DTYPE_FLOAT32 446: "text.445_of_512" NUMERICAL mean:-0.00269593 min:-0.0824544 max:0.0926119 sd:0.0370001 dtype:DTYPE_FLOAT32 447: "text.446_of_512" NUMERICAL mean:0.015445 min:-0.0938269 max:0.0821593 sd:0.032405 dtype:DTYPE_FLOAT32 448: "text.447_of_512" NUMERICAL mean:-0.00200815 min:-0.0900273 max:0.0887034 sd:0.0375671 dtype:DTYPE_FLOAT32 449: "text.448_of_512" NUMERICAL mean:-0.00702704 min:-0.0910169 max:0.0897229 sd:0.0376343 dtype:DTYPE_FLOAT32 450: "text.449_of_512" NUMERICAL mean:0.0235369 min:-0.0798573 max:0.0910766 sd:0.0326688 dtype:DTYPE_FLOAT32 451: "text.450_of_512" NUMERICAL mean:-0.00924271 min:-0.0965331 max:0.0841149 sd:0.0392387 dtype:DTYPE_FLOAT32 452: "text.451_of_512" NUMERICAL mean:-0.0159282 min:-0.0900897 max:0.0855076 sd:0.035355 dtype:DTYPE_FLOAT32 453: "text.452_of_512" NUMERICAL mean:0.00869069 min:-0.073467 max:0.0844256 sd:0.0345292 dtype:DTYPE_FLOAT32 454: "text.453_of_512" NUMERICAL mean:0.0329876 min:-0.0882511 max:0.0865172 sd:0.0307093 dtype:DTYPE_FLOAT32 455: "text.454_of_512" NUMERICAL mean:-0.0305705 min:-0.0977973 max:0.0805249 sd:0.0315637 dtype:DTYPE_FLOAT32 456: "text.455_of_512" NUMERICAL mean:-0.043483 min:-0.093159 max:0.100654 sd:0.0345805 dtype:DTYPE_FLOAT32 457: "text.456_of_512" NUMERICAL mean:0.00927735 min:-0.107284 max:0.0896101 sd:0.0400691 dtype:DTYPE_FLOAT32 458: "text.457_of_512" NUMERICAL mean:0.0309873 min:-0.0966208 max:0.085311 sd:0.0322956 dtype:DTYPE_FLOAT32 459: "text.458_of_512" NUMERICAL mean:-0.0592233 min:-0.113983 max:0.0753078 sd:0.020848 dtype:DTYPE_FLOAT32 460: "text.459_of_512" NUMERICAL mean:-5.20169e-05 min:-0.0815561 max:0.0879604 sd:0.035764 dtype:DTYPE_FLOAT32 461: "text.460_of_512" NUMERICAL mean:-0.0255917 min:-0.105131 max:0.0741838 sd:0.0376767 dtype:DTYPE_FLOAT32 462: "text.461_of_512" NUMERICAL mean:0.0464927 min:-0.0876184 max:0.0943474 sd:0.0249704 dtype:DTYPE_FLOAT32 463: "text.462_of_512" NUMERICAL mean:0.00871853 min:-0.0926539 max:0.0871729 sd:0.0369573 dtype:DTYPE_FLOAT32 464: "text.463_of_512" NUMERICAL mean:0.0228398 min:-0.102781 max:0.0935057 sd:0.0393205 dtype:DTYPE_FLOAT32 465: "text.464_of_512" NUMERICAL mean:0.00286421 min:-0.0814077 max:0.101914 sd:0.0434828 dtype:DTYPE_FLOAT32 466: "text.465_of_512" NUMERICAL mean:0.0292556 min:-0.0838276 max:0.0820808 sd:0.0300237 dtype:DTYPE_FLOAT32 467: "text.466_of_512" NUMERICAL mean:0.00325644 min:-0.0918644 max:0.0857691 sd:0.0353731 dtype:DTYPE_FLOAT32 468: "text.467_of_512" NUMERICAL mean:0.00708703 min:-0.0926971 max:0.0799828 sd:0.0344336 dtype:DTYPE_FLOAT32 469: "text.468_of_512" NUMERICAL mean:0.00258697 min:-0.0881227 max:0.0879532 sd:0.0367695 dtype:DTYPE_FLOAT32 470: "text.469_of_512" NUMERICAL mean:0.0153641 min:-0.0720187 max:0.0888214 sd:0.0315247 dtype:DTYPE_FLOAT32 471: "text.470_of_512" NUMERICAL mean:-0.0132175 min:-0.0884372 max:0.0887741 sd:0.0398849 dtype:DTYPE_FLOAT32 472: "text.471_of_512" NUMERICAL mean:0.0283167 min:-0.0787876 max:0.0958703 sd:0.0339658 dtype:DTYPE_FLOAT32 473: "text.472_of_512" NUMERICAL mean:-0.00888963 min:-0.0856551 max:0.0953346 sd:0.0377869 dtype:DTYPE_FLOAT32 474: "text.473_of_512" NUMERICAL mean:-0.00262449 min:-0.0851525 max:0.0867075 sd:0.0353713 dtype:DTYPE_FLOAT32 475: "text.474_of_512" NUMERICAL mean:-0.0234516 min:-0.090224 max:0.0760985 sd:0.0289638 dtype:DTYPE_FLOAT32 476: "text.475_of_512" NUMERICAL mean:0.00604867 min:-0.0872303 max:0.0876132 sd:0.0382356 dtype:DTYPE_FLOAT32 477: "text.476_of_512" NUMERICAL mean:-0.00365237 min:-0.0938913 max:0.0851847 sd:0.0384193 dtype:DTYPE_FLOAT32 478: "text.477_of_512" NUMERICAL mean:-0.0224806 min:-0.0898023 max:0.0856496 sd:0.0361166 dtype:DTYPE_FLOAT32 479: "text.478_of_512" NUMERICAL mean:0.0692429 min:-0.0674553 max:0.126405 sd:0.0170738 dtype:DTYPE_FLOAT32 480: "text.479_of_512" NUMERICAL mean:-0.03593 min:-0.100611 max:0.0843861 sd:0.0365605 dtype:DTYPE_FLOAT32 481: "text.480_of_512" NUMERICAL mean:0.0051222 min:-0.0911619 max:0.0917946 sd:0.0381134 dtype:DTYPE_FLOAT32 482: "text.481_of_512" NUMERICAL mean:-0.00217864 min:-0.097742 max:0.0793843 sd:0.032152 dtype:DTYPE_FLOAT32 483: "text.482_of_512" NUMERICAL mean:-0.0186551 min:-0.0844438 max:0.0839448 sd:0.0315299 dtype:DTYPE_FLOAT32 484: "text.483_of_512" NUMERICAL mean:-0.0726262 min:-0.106575 max:0.0346921 sd:0.0089769 dtype:DTYPE_FLOAT32 485: "text.484_of_512" NUMERICAL mean:0.0218913 min:-0.0899297 max:0.0855874 sd:0.0364844 dtype:DTYPE_FLOAT32 486: "text.485_of_512" NUMERICAL mean:-0.0317 min:-0.0915854 max:0.081424 sd:0.0335331 dtype:DTYPE_FLOAT32 487: "text.486_of_512" NUMERICAL mean:0.0166038 min:-0.0937347 max:0.105147 sd:0.0349845 dtype:DTYPE_FLOAT32 488: "text.487_of_512" NUMERICAL mean:0.062243 min:-0.0958989 max:0.116609 sd:0.0354013 dtype:DTYPE_FLOAT32 489: "text.488_of_512" NUMERICAL mean:-0.00663054 min:-0.0901465 max:0.0806198 sd:0.0335516 dtype:DTYPE_FLOAT32 490: "text.489_of_512" NUMERICAL mean:0.000481812 min:-0.0856375 max:0.0848821 sd:0.0383612 dtype:DTYPE_FLOAT32 491: "text.490_of_512" NUMERICAL mean:0.0130283 min:-0.0820245 max:0.0937727 sd:0.035001 dtype:DTYPE_FLOAT32 492: "text.491_of_512" NUMERICAL mean:0.00648594 min:-0.0897108 max:0.0934784 sd:0.0351927 dtype:DTYPE_FLOAT32 493: "text.492_of_512" NUMERICAL mean:-0.0339618 min:-0.0869285 max:0.0855316 sd:0.0337819 dtype:DTYPE_FLOAT32 494: "text.493_of_512" NUMERICAL mean:0.019294 min:-0.07954 max:0.0884682 sd:0.0299775 dtype:DTYPE_FLOAT32 495: "text.494_of_512" NUMERICAL mean:0.0146976 min:-0.0782096 max:0.0908863 sd:0.0332566 dtype:DTYPE_FLOAT32 496: "text.495_of_512" NUMERICAL mean:0.000877017 min:-0.0916627 max:0.0893738 sd:0.0353427 dtype:DTYPE_FLOAT32 497: "text.496_of_512" NUMERICAL mean:-0.00545176 min:-0.0875419 max:0.0954239 sd:0.0382569 dtype:DTYPE_FLOAT32 498: "text.497_of_512" NUMERICAL mean:-0.00126044 min:-0.083081 max:0.0876477 sd:0.0370372 dtype:DTYPE_FLOAT32 499: "text.498_of_512" NUMERICAL mean:0.00691261 min:-0.081251 max:0.0837367 sd:0.0311204 dtype:DTYPE_FLOAT32 500: "text.499_of_512" NUMERICAL mean:-0.0219051 min:-0.0896912 max:0.0858947 sd:0.0349078 dtype:DTYPE_FLOAT32 501: "text.500_of_512" NUMERICAL mean:-0.0132117 min:-0.0867502 max:0.0888057 sd:0.0370112 dtype:DTYPE_FLOAT32 502: "text.501_of_512" NUMERICAL mean:-0.0621911 min:-0.106201 max:0.101876 sd:0.0250442 dtype:DTYPE_FLOAT32 503: "text.502_of_512" NUMERICAL mean:-0.0186272 min:-0.105629 max:0.0821297 sd:0.0343729 dtype:DTYPE_FLOAT32 504: "text.503_of_512" NUMERICAL mean:-0.0729121 min:-0.102538 max:0.0918324 sd:0.0116754 dtype:DTYPE_FLOAT32 505: "text.504_of_512" NUMERICAL mean:0.0169701 min:-0.0895194 max:0.097876 sd:0.0349639 dtype:DTYPE_FLOAT32 506: "text.505_of_512" NUMERICAL mean:-0.0090915 min:-0.0935523 max:0.0770736 sd:0.0349964 dtype:DTYPE_FLOAT32 507: "text.506_of_512" NUMERICAL mean:-0.00631471 min:-0.0873942 max:0.0917989 sd:0.0358899 dtype:DTYPE_FLOAT32 508: "text.507_of_512" NUMERICAL mean:0.0061374 min:-0.0830885 max:0.10424 sd:0.0373206 dtype:DTYPE_FLOAT32 509: "text.508_of_512" NUMERICAL mean:0.0113484 min:-0.0807295 max:0.0840633 sd:0.0339968 dtype:DTYPE_FLOAT32 510: "text.509_of_512" NUMERICAL mean:-0.000169981 min:-0.0808745 max:0.0819429 sd:0.0343803 dtype:DTYPE_FLOAT32 511: "text.510_of_512" NUMERICAL mean:0.0724568 min:-0.0795371 max:0.117238 sd:0.0176485 dtype:DTYPE_FLOAT32 512: "text.511_of_512" NUMERICAL mean:0.0159748 min:-0.0964746 max:0.0833914 sd:0.0401069 dtype:DTYPE_FLOAT32 CATEGORICAL: 1 (0.194932%) 0: "label" CATEGORICAL has-dict vocab-size:3 zero-ood-items most-frequent:"0" 12500 (50%) 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): 0.649996 Accuracy: 0.863545 CI95[W][0 1] ErrorRate: : 0.136455 Confusion Table: truth\prediction 0 1 0 1070 164 1 174 1069 Total: 2477
Variable importances measure the importance of an input feature for a model.
1. "text.111_of_512" 0.186278 ################ 2. "text.384_of_512" 0.181487 ########### 3. "text.171_of_512" 0.181331 ########### 4. "text.022_of_512" 0.178618 ######## 5. "text.305_of_512" 0.177880 ####### 6. "text.459_of_512" 0.177238 ####### 7. "text.386_of_512" 0.175864 ##### 8. "text.408_of_512" 0.175837 ##### 9. "text.307_of_512" 0.175136 ##### 10. "text.182_of_512" 0.175062 ##### 11. "text.353_of_512" 0.174753 #### 12. "text.308_of_512" 0.174689 #### 13. "text.302_of_512" 0.174182 #### 14. "text.132_of_512" 0.173670 ### 15. "text.281_of_512" 0.173650 ### 16. "text.006_of_512" 0.173577 ### 17. "text.402_of_512" 0.173513 ### 18. "text.023_of_512" 0.173490 ### 19. "text.237_of_512" 0.173229 ### 20. "text.284_of_512" 0.173132 ### 21. "text.245_of_512" 0.173120 ### 22. "text.419_of_512" 0.173038 ### 23. "text.361_of_512" 0.172992 ### 24. "text.167_of_512" 0.172913 ### 25. "text.352_of_512" 0.172859 ## 26. "text.276_of_512" 0.172856 ## 27. "text.056_of_512" 0.172752 ## 28. "text.173_of_512" 0.172657 ## 29. "text.201_of_512" 0.172582 ## 30. "text.038_of_512" 0.172572 ## 31. "text.300_of_512" 0.172337 ## 32. "text.090_of_512" 0.172299 ## 33. "text.232_of_512" 0.172254 ## 34. "text.065_of_512" 0.172242 ## 35. "text.342_of_512" 0.172230 ## 36. "text.155_of_512" 0.172213 ## 37. "text.079_of_512" 0.172180 ## 38. "text.251_of_512" 0.172155 ## 39. "text.064_of_512" 0.172139 ## 40. "text.041_of_512" 0.172060 ## 41. "text.482_of_512" 0.172048 ## 42. "text.157_of_512" 0.171982 ## 43. "text.082_of_512" 0.171889 ## 44. "text.017_of_512" 0.171882 ## 45. "text.452_of_512" 0.171863 ## 46. "text.439_of_512" 0.171847 # 47. "text.078_of_512" 0.171789 # 48. "text.198_of_512" 0.171743 # 49. "text.328_of_512" 0.171734 # 50. "text.046_of_512" 0.171713 # 51. "text.045_of_512" 0.171687 # 52. "text.035_of_512" 0.171665 # 53. "text.505_of_512" 0.171613 # 54. "text.200_of_512" 0.171605 # 55. "text.190_of_512" 0.171530 # 56. "text.068_of_512" 0.171527 # 57. "text.407_of_512" 0.171516 # 58. "text.314_of_512" 0.171407 # 59. "text.293_of_512" 0.171403 # 60. "text.471_of_512" 0.171365 # 61. "text.063_of_512" 0.171362 # 62. "text.140_of_512" 0.171356 # 63. "text.435_of_512" 0.171334 # 64. "text.443_of_512" 0.171316 # 65. "text.199_of_512" 0.171314 # 66. "text.168_of_512" 0.171303 # 67. "text.350_of_512" 0.171278 # 68. "text.485_of_512" 0.171273 # 69. "text.400_of_512" 0.171269 # 70. "text.379_of_512" 0.171257 # 71. "text.388_of_512" 0.171223 # 72. "text.120_of_512" 0.171152 # 73. "text.043_of_512" 0.171148 # 74. "text.195_of_512" 0.171125 # 75. "text.126_of_512" 0.171111 # 76. "text.192_of_512" 0.171102 # 77. "text.381_of_512" 0.171092 # 78. "text.349_of_512" 0.171089 # 79. "text.150_of_512" 0.171083 # 80. "text.092_of_512" 0.171076 # 81. "text.240_of_512" 0.171070 # 82. "text.204_of_512" 0.171055 # 83. "text.294_of_512" 0.171029 # 84. "text.496_of_512" 0.171023 # 85. "text.133_of_512" 0.170987 # 86. "text.444_of_512" 0.170980 # 87. "text.259_of_512" 0.170964 # 88. "text.370_of_512" 0.170943 # 89. "text.449_of_512" 0.170939 # 90. "text.306_of_512" 0.170937 # 91. "text.212_of_512" 0.170931 # 92. "text.365_of_512" 0.170931 # 93. "text.239_of_512" 0.170923 # 94. "text.442_of_512" 0.170912 # 95. "text.324_of_512" 0.170890 # 96. "text.194_of_512" 0.170874 # 97. "text.317_of_512" 0.170853 # 98. "text.354_of_512" 0.170842 # 99. "text.071_of_512" 0.170841 # 100. "text.269_of_512" 0.170840 # 101. "text.313_of_512" 0.170810 102. "text.179_of_512" 0.170805 103. "text.032_of_512" 0.170805 104. "text.048_of_512" 0.170804 105. "text.188_of_512" 0.170796 106. "text.463_of_512" 0.170795 107. "text.169_of_512" 0.170791 108. "text.364_of_512" 0.170791 109. "text.446_of_512" 0.170780 110. "text.403_of_512" 0.170771 111. "text.415_of_512" 0.170762 112. "text.273_of_512" 0.170756 113. "text.099_of_512" 0.170752 114. "text.346_of_512" 0.170748 115. "text.067_of_512" 0.170733 116. "text.027_of_512" 0.170721 117. "text.238_of_512" 0.170713 118. "text.069_of_512" 0.170710 119. "text.398_of_512" 0.170702 120. "text.351_of_512" 0.170698 121. "text.405_of_512" 0.170696 122. "text.373_of_512" 0.170694 123. "text.425_of_512" 0.170690 124. "text.142_of_512" 0.170681 125. "text.458_of_512" 0.170665 126. "text.296_of_512" 0.170657 127. "text.434_of_512" 0.170656 128. "text.209_of_512" 0.170652 129. "text.031_of_512" 0.170646 130. "text.051_of_512" 0.170643 131. "text.376_of_512" 0.170634 132. "text.494_of_512" 0.170632 133. "text.375_of_512" 0.170626 134. "text.219_of_512" 0.170621 135. "text.176_of_512" 0.170615 136. "text.410_of_512" 0.170598 137. "text.118_of_512" 0.170594 138. "text.217_of_512" 0.170589 139. "text.322_of_512" 0.170580 140. "text.014_of_512" 0.170561 141. "text.500_of_512" 0.170558 142. "text.476_of_512" 0.170554 143. "text.144_of_512" 0.170545 144. "text.246_of_512" 0.170524 145. "text.348_of_512" 0.170506 146. "text.431_of_512" 0.170504 147. "text.059_of_512" 0.170504 148. "text.049_of_512" 0.170502 149. "text.331_of_512" 0.170502 150. "text.344_of_512" 0.170496 151. "text.123_of_512" 0.170491 152. "text.101_of_512" 0.170490 153. "text.089_of_512" 0.170486 154. "text.267_of_512" 0.170483 155. "text.007_of_512" 0.170477 156. "text.347_of_512" 0.170476 157. "text.205_of_512" 0.170472 158. "text.492_of_512" 0.170467 159. "text.382_of_512" 0.170463 160. "text.005_of_512" 0.170453 161. "text.196_of_512" 0.170438 162. "text.076_of_512" 0.170436 163. "text.263_of_512" 0.170428 164. "text.499_of_512" 0.170426 165. "text.436_of_512" 0.170426 166. "text.244_of_512" 0.170414 167. "text.037_of_512" 0.170410 168. "text.149_of_512" 0.170405 169. "text.450_of_512" 0.170404 170. "text.026_of_512" 0.170395 171. "text.272_of_512" 0.170392 172. "text.104_of_512" 0.170384 173. "text.395_of_512" 0.170381 174. "text.030_of_512" 0.170380 175. "text.319_of_512" 0.170374 176. "text.234_of_512" 0.170373 177. "text.297_of_512" 0.170372 178. "text.042_of_512" 0.170371 179. "text.117_of_512" 0.170370 180. "text.093_of_512" 0.170358 181. "text.151_of_512" 0.170355 182. "text.184_of_512" 0.170353 183. "text.424_of_512" 0.170352 184. "text.073_of_512" 0.170349 185. "text.085_of_512" 0.170347 186. "text.044_of_512" 0.170346 187. "text.392_of_512" 0.170346 188. "text.136_of_512" 0.170345 189. "text.224_of_512" 0.170340 190. "text.202_of_512" 0.170339 191. "text.429_of_512" 0.170328 192. "text.369_of_512" 0.170326 193. "text.012_of_512" 0.170322 194. "text.057_of_512" 0.170317 195. "text.274_of_512" 0.170316 196. "text.110_of_512" 0.170315 197. "text.102_of_512" 0.170312 198. "text.172_of_512" 0.170311 199. "text.261_of_512" 0.170307 200. "text.100_of_512" 0.170300 201. "text.241_of_512" 0.170294 202. "text.454_of_512" 0.170292 203. "text.137_of_512" 0.170289 204. "text.428_of_512" 0.170285 205. "text.220_of_512" 0.170283 206. "text.034_of_512" 0.170283 207. "text.165_of_512" 0.170282 208. "text.448_of_512" 0.170281 209. "text.484_of_512" 0.170277 210. "text.467_of_512" 0.170277 211. "text.295_of_512" 0.170273 212. "text.208_of_512" 0.170273 213. "text.095_of_512" 0.170270 214. "text.412_of_512" 0.170270 215. "text.236_of_512" 0.170263 216. "text.039_of_512" 0.170263 217. "text.451_of_512" 0.170250 218. "text.024_of_512" 0.170248 219. "text.112_of_512" 0.170247 220. "text.420_of_512" 0.170245 221. "text.478_of_512" 0.170237 222. "text.456_of_512" 0.170237 223. "text.086_of_512" 0.170228 224. "text.158_of_512" 0.170227 225. "text.124_of_512" 0.170225 226. "text.009_of_512" 0.170219 227. "text.283_of_512" 0.170214 228. "text.290_of_512" 0.170213 229. "text.229_of_512" 0.170212 230. "text.040_of_512" 0.170211 231. "text.228_of_512" 0.170211 232. "text.181_of_512" 0.170209 233. "text.357_of_512" 0.170206 234. "text.162_of_512" 0.170201 235. "text.258_of_512" 0.170201 236. "text.131_of_512" 0.170198 237. "text.474_of_512" 0.170196 238. "text.286_of_512" 0.170196 239. "text.413_of_512" 0.170192 240. "text.203_of_512" 0.170190 241. "text.070_of_512" 0.170188 242. "text.047_of_512" 0.170187 243. "text.215_of_512" 0.170180 244. "text.054_of_512" 0.170180 245. "text.166_of_512" 0.170178 246. "text.481_of_512" 0.170178 247. "text.013_of_512" 0.170174 248. "text.404_of_512" 0.170174 249. "text.187_of_512" 0.170172 250. "text.282_of_512" 0.170172 251. "text.197_of_512" 0.170171 252. "text.316_of_512" 0.170169 253. "text.177_of_512" 0.170169 254. "text.311_of_512" 0.170168 255. "text.389_of_512" 0.170164 256. "text.011_of_512" 0.170162 257. "text.001_of_512" 0.170162 258. "text.417_of_512" 0.170162 259. "text.021_of_512" 0.170158 260. "text.113_of_512" 0.170152 261. "text.265_of_512" 0.170150 262. "text.416_of_512" 0.170149 263. "text.260_of_512" 0.170149 264. "text.135_of_512" 0.170147 265. "text.489_of_512" 0.170144 266. "text.509_of_512" 0.170143 267. "text.360_of_512" 0.170140 268. "text.114_of_512" 0.170137 269. "text.066_of_512" 0.170135 270. "text.119_of_512" 0.170134 271. "text.138_of_512" 0.170133 272. "text.466_of_512" 0.170133 273. "text.028_of_512" 0.170131 274. "text.002_of_512" 0.170131 275. "text.469_of_512" 0.170130 276. "text.122_of_512" 0.170129 277. "text.472_of_512" 0.170129 278. "text.148_of_512" 0.170127 279. "text.227_of_512" 0.170125 280. "text.337_of_512" 0.170120 281. "text.091_of_512" 0.170118 282. "text.097_of_512" 0.170117 283. "text.145_of_512" 0.170116 284. "text.501_of_512" 0.170112 285. "text.355_of_512" 0.170108 286. "text.154_of_512" 0.170106 287. "text.327_of_512" 0.170102 288. "text.053_of_512" 0.170101 289. "text.159_of_512" 0.170097 290. "text.189_of_512" 0.170096 291. "text.473_of_512" 0.170095 292. "text.374_of_512" 0.170093 293. "text.106_of_512" 0.170092 294. "text.108_of_512" 0.170090 295. "text.498_of_512" 0.170088 296. "text.371_of_512" 0.170087 297. "text.210_of_512" 0.170085 298. "text.183_of_512" 0.170084 299. "text.256_of_512" 0.170083 300. "text.074_of_512" 0.170082 301. "text.422_of_512" 0.170081 302. "text.502_of_512" 0.170079 303. "text.487_of_512" 0.170078 304. "text.391_of_512" 0.170078 305. "text.152_of_512" 0.170078 306. "text.312_of_512" 0.170078 307. "text.186_of_512" 0.170076 308. "text.275_of_512" 0.170076 309. "text.287_of_512" 0.170074 310. "text.072_of_512" 0.170074 311. "text.214_of_512" 0.170070 312. "text.363_of_512" 0.170069 313. "text.226_of_512" 0.170067 314. "text.061_of_512" 0.170067 315. "text.242_of_512" 0.170064 316. "text.393_of_512" 0.170063 317. "text.125_of_512" 0.170063 318. "text.504_of_512" 0.170061 319. "text.475_of_512" 0.170059 320. "text.303_of_512" 0.170058 321. "text.087_of_512" 0.170058 322. "text.421_of_512" 0.170058 323. "text.170_of_512" 0.170054 324. "text.289_of_512" 0.170053 325. "text.465_of_512" 0.170052 326. "text.252_of_512" 0.170051 327. "text.008_of_512" 0.170049 328. "text.058_of_512" 0.170044 329. "text.418_of_512" 0.170044 330. "text.310_of_512" 0.170039 331. "text.096_of_512" 0.170036 332. "text.479_of_512" 0.170034 333. "text.341_of_512" 0.170033 334. "text.396_of_512" 0.170032 335. "text.185_of_512" 0.170030 336. "text.015_of_512" 0.170030 337. "text.329_of_512" 0.170028 338. "text.247_of_512" 0.170027 339. "text.430_of_512" 0.170026 340. "text.461_of_512" 0.170025 341. "text.019_of_512" 0.170025 342. "text.483_of_512" 0.170024 343. "text.339_of_512" 0.170023 344. "text.397_of_512" 0.170020 345. "text.262_of_512" 0.170013 346. "text.018_of_512" 0.170010 347. "text.423_of_512" 0.170008 348. "text.163_of_512" 0.170008 349. "text.139_of_512" 0.170008 350. "text.372_of_512" 0.170007 351. "text.401_of_512" 0.170004 352. "text.304_of_512" 0.170002 353. "text.367_of_512" 0.170000 354. "text.161_of_512" 0.170000 355. "text.468_of_512" 0.169999 356. "text.390_of_512" 0.169999 357. "text.174_of_512" 0.169998 358. "text.243_of_512" 0.169997 359. "text.315_of_512" 0.169996 360. "text.511_of_512" 0.169993 361. "text.280_of_512" 0.169991 362. "text.301_of_512" 0.169991 363. "text.083_of_512" 0.169990 364. "text.025_of_512" 0.169990 365. "text.292_of_512" 0.169990 366. "text.207_of_512" 0.169987 367. "text.321_of_512" 0.169981 368. "text.491_of_512" 0.169981 369. "text.075_of_512" 0.169980 370. "text.225_of_512" 0.169978 371. "text.077_of_512" 0.169977 372. "text.129_of_512" 0.169975 373. "text.414_of_512" 0.169975 374. "text.249_of_512" 0.169973 375. "text.221_of_512" 0.169973 376. "text.334_of_512" 0.169972 377. "text.216_of_512" 0.169971 378. "text.175_of_512" 0.169970 379. "text.335_of_512" 0.169970 380. "text.426_of_512" 0.169965 381. "text.147_of_512" 0.169964 382. "text.180_of_512" 0.169964 383. "text.105_of_512" 0.169962 384. "text.368_of_512" 0.169961 385. "text.488_of_512" 0.169960 386. "text.121_of_512" 0.169959 387. "text.178_of_512" 0.169959 388. "text.411_of_512" 0.169957 389. "text.191_of_512" 0.169956 390. "text.213_of_512" 0.169956 391. "text.211_of_512" 0.169955 392. "text.495_of_512" 0.169954 393. "text.103_of_512" 0.169952 394. "text.143_of_512" 0.169951 395. "text.291_of_512" 0.169951 396. "text.359_of_512" 0.169947 397. "text.218_of_512" 0.169946 398. "text.156_of_512" 0.169945 399. "text.480_of_512" 0.169944 400. "text.248_of_512" 0.169941 401. "text.033_of_512" 0.169941 402. "text.497_of_512" 0.169940 403. "text.356_of_512" 0.169938 404. "text.164_of_512" 0.169938 405. "text.409_of_512" 0.169936 406. "text.010_of_512" 0.169936 407. "text.264_of_512" 0.169935 408. "text.130_of_512" 0.169935 409. "text.358_of_512" 0.169935 410. "text.000_of_512" 0.169935 411. "text.433_of_512" 0.169934 412. "text.134_of_512" 0.169933 413. "text.107_of_512" 0.169933 414. "text.222_of_512" 0.169931 415. "text.486_of_512" 0.169928 416. "text.050_of_512" 0.169926 417. "text.387_of_512" 0.169926 418. "text.464_of_512" 0.169925 419. "text.029_of_512" 0.169925 420. "text.116_of_512" 0.169925 421. "text.055_of_512" 0.169922 422. "text.362_of_512" 0.169921 423. "text.457_of_512" 0.169920 424. "text.253_of_512" 0.169920 425. "text.160_of_512" 0.169918 426. "text.004_of_512" 0.169917 427. "text.016_of_512" 0.169917 428. "text.462_of_512" 0.169916 429. "text.268_of_512" 0.169914 430. "text.115_of_512" 0.169913 431. "text.230_of_512" 0.169912 432. "text.366_of_512" 0.169912 433. "text.003_of_512" 0.169909 434. "text.146_of_512" 0.169909 435. "text.141_of_512" 0.169908 436. "text.323_of_512" 0.169906 437. "text.127_of_512" 0.169906 438. "text.493_of_512" 0.169905 439. "text.432_of_512" 0.169903 440. "text.445_of_512" 0.169902 441. "text.343_of_512" 0.169898 442. "text.345_of_512" 0.169893 443. "text.223_of_512" 0.169892 444. "text.508_of_512" 0.169890 445. "text.437_of_512" 0.169889 446. "text.060_of_512" 0.169889 447. "text.333_of_512" 0.169888 448. "text.020_of_512" 0.169888 449. "text.094_of_512" 0.169887 450. "text.299_of_512" 0.169886 451. "text.490_of_512" 0.169885 452. "text.330_of_512" 0.169884 453. "text.427_of_512" 0.169882 454. "text.506_of_512" 0.169881 455. "text.380_of_512" 0.169880 456. "text.441_of_512" 0.169880 457. "text.109_of_512" 0.169878 458. "text.279_of_512" 0.169877 459. "text.255_of_512" 0.169877 460. "text.062_of_512" 0.169876 461. "text.326_of_512" 0.169875 462. "text.084_of_512" 0.169875 463. "text.447_of_512" 0.169873 464. "text.235_of_512" 0.169872 465. "text.052_of_512" 0.169872 466. "text.036_of_512" 0.169870 467. "text.271_of_512" 0.169869 468. "text.266_of_512" 0.169868 469. "text.332_of_512" 0.169866 470. "text.081_of_512" 0.169864 471. "text.320_of_512" 0.169863 472. "text.394_of_512" 0.169862 473. "text.298_of_512" 0.169861 474. "text.460_of_512" 0.169861 475. "text.325_of_512" 0.169860 476. "text.477_of_512" 0.169857 477. "text.080_of_512" 0.169855 478. "text.277_of_512" 0.169854 479. "text.336_of_512" 0.169850 480. "text.503_of_512" 0.169848 481. "text.233_of_512" 0.169848 482. "text.088_of_512" 0.169846 483. "text.440_of_512" 0.169845 484. "text.378_of_512" 0.169845 485. "text.309_of_512" 0.169844 486. "text.128_of_512" 0.169843 487. "text.193_of_512" 0.169843 488. "text.385_of_512" 0.169842 489. "text.438_of_512" 0.169842 490. "text.338_of_512" 0.169837 491. "text.231_of_512" 0.169836 492. "text.507_of_512" 0.169835 493. "text.377_of_512" 0.169834 494. "text.254_of_512" 0.169831 495. "text.470_of_512" 0.169828 496. "text.257_of_512" 0.169826 497. "text.453_of_512" 0.169825 498. "text.340_of_512" 0.169825 499. "text.153_of_512" 0.169823 500. "text.318_of_512" 0.169822 501. "text.206_of_512" 0.169820 502. "text.406_of_512" 0.169818 503. "text.455_of_512" 0.169818 504. "text.098_of_512" 0.169818 505. "text.383_of_512" 0.169818 506. "text.399_of_512" 0.169816 507. "text.278_of_512" 0.169815 508. "text.285_of_512" 0.169814 509. "text.288_of_512" 0.169793 510. "text.270_of_512" 0.169792 511. "text.250_of_512" 0.169791
1. "text.111_of_512" 25.000000 ################ 2. "text.384_of_512" 17.000000 ########## 3. "text.171_of_512" 16.000000 ########## 4. "text.305_of_512" 13.000000 ######## 5. "text.459_of_512" 9.000000 ##### 6. "text.022_of_512" 8.000000 #### 7. "text.353_of_512" 8.000000 #### 8. "text.182_of_512" 6.000000 ### 9. "text.307_of_512" 6.000000 ### 10. "text.386_of_512" 6.000000 ### 11. "text.006_of_512" 5.000000 ## 12. "text.023_of_512" 5.000000 ## 13. "text.281_of_512" 5.000000 ## 14. "text.284_of_512" 5.000000 ## 15. "text.302_of_512" 5.000000 ## 16. "text.308_of_512" 5.000000 ## 17. "text.361_of_512" 5.000000 ## 18. "text.402_of_512" 5.000000 ## 19. "text.408_of_512" 5.000000 ## 20. "text.419_of_512" 5.000000 ## 21. "text.056_of_512" 4.000000 ## 22. "text.064_of_512" 4.000000 ## 23. "text.167_of_512" 4.000000 ## 24. "text.201_of_512" 4.000000 ## 25. "text.237_of_512" 4.000000 ## 26. "text.038_of_512" 3.000000 # 27. "text.078_of_512" 3.000000 # 28. "text.079_of_512" 3.000000 # 29. "text.155_of_512" 3.000000 # 30. "text.157_of_512" 3.000000 # 31. "text.173_of_512" 3.000000 # 32. "text.200_of_512" 3.000000 # 33. "text.232_of_512" 3.000000 # 34. "text.245_of_512" 3.000000 # 35. "text.276_of_512" 3.000000 # 36. "text.352_of_512" 3.000000 # 37. "text.505_of_512" 3.000000 # 38. "text.017_of_512" 2.000000 39. "text.035_of_512" 2.000000 40. "text.041_of_512" 2.000000 41. "text.045_of_512" 2.000000 42. "text.090_of_512" 2.000000 43. "text.120_of_512" 2.000000 44. "text.132_of_512" 2.000000 45. "text.133_of_512" 2.000000 46. "text.199_of_512" 2.000000 47. "text.204_of_512" 2.000000 48. "text.251_of_512" 2.000000 49. "text.293_of_512" 2.000000 50. "text.294_of_512" 2.000000 51. "text.314_of_512" 2.000000 52. "text.328_of_512" 2.000000 53. "text.349_of_512" 2.000000 54. "text.370_of_512" 2.000000 55. "text.439_of_512" 2.000000 56. "text.444_of_512" 2.000000 57. "text.452_of_512" 2.000000 58. "text.485_of_512" 2.000000 59. "text.496_of_512" 2.000000 60. "text.007_of_512" 1.000000 61. "text.014_of_512" 1.000000 62. "text.027_of_512" 1.000000 63. "text.032_of_512" 1.000000 64. "text.042_of_512" 1.000000 65. "text.043_of_512" 1.000000 66. "text.063_of_512" 1.000000 67. "text.068_of_512" 1.000000 68. "text.082_of_512" 1.000000 69. "text.126_of_512" 1.000000 70. "text.150_of_512" 1.000000 71. "text.169_of_512" 1.000000 72. "text.179_of_512" 1.000000 73. "text.190_of_512" 1.000000 74. "text.195_of_512" 1.000000 75. "text.198_of_512" 1.000000 76. "text.209_of_512" 1.000000 77. "text.234_of_512" 1.000000 78. "text.238_of_512" 1.000000 79. "text.246_of_512" 1.000000 80. "text.259_of_512" 1.000000 81. "text.267_of_512" 1.000000 82. "text.290_of_512" 1.000000 83. "text.300_of_512" 1.000000 84. "text.342_of_512" 1.000000 85. "text.350_of_512" 1.000000 86. "text.381_of_512" 1.000000 87. "text.395_of_512" 1.000000 88. "text.400_of_512" 1.000000 89. "text.407_of_512" 1.000000 90. "text.410_of_512" 1.000000 91. "text.415_of_512" 1.000000 92. "text.446_of_512" 1.000000 93. "text.471_of_512" 1.000000 94. "text.482_of_512" 1.000000 95. "text.500_of_512" 1.000000
1. "text.111_of_512" 104.000000 ################ 2. "text.022_of_512" 91.000000 ############# 3. "text.171_of_512" 89.000000 ############# 4. "text.384_of_512" 81.000000 ############ 5. "text.408_of_512" 65.000000 ######### 6. "text.182_of_512" 63.000000 ######### 7. "text.065_of_512" 61.000000 ######### 8. "text.459_of_512" 59.000000 ######## 9. "text.245_of_512" 58.000000 ######## 10. "text.386_of_512" 56.000000 ######## 11. "text.300_of_512" 53.000000 ######## 12. "text.132_of_512" 49.000000 ####### 13. "text.237_of_512" 49.000000 ####### 14. "text.308_of_512" 48.000000 ####### 15. "text.046_of_512" 47.000000 ####### 16. "text.302_of_512" 47.000000 ####### 17. "text.251_of_512" 45.000000 ###### 18. "text.006_of_512" 44.000000 ###### 19. "text.281_of_512" 44.000000 ###### 20. "text.305_of_512" 44.000000 ###### 21. "text.276_of_512" 43.000000 ###### 22. "text.342_of_512" 43.000000 ###### 23. "text.307_of_512" 40.000000 ##### 24. "text.056_of_512" 39.000000 ##### 25. "text.419_of_512" 38.000000 ##### 26. "text.482_of_512" 38.000000 ##### 27. "text.190_of_512" 37.000000 ##### 28. "text.198_of_512" 37.000000 ##### 29. "text.353_of_512" 37.000000 ##### 30. "text.082_of_512" 36.000000 ##### 31. "text.350_of_512" 36.000000 ##### 32. "text.194_of_512" 35.000000 ##### 33. "text.452_of_512" 35.000000 ##### 34. "text.168_of_512" 34.000000 ##### 35. "text.435_of_512" 34.000000 ##### 36. "text.463_of_512" 34.000000 ##### 37. "text.045_of_512" 33.000000 #### 38. "text.140_of_512" 33.000000 #### 39. "text.240_of_512" 33.000000 #### 40. "text.043_of_512" 32.000000 #### 41. "text.173_of_512" 32.000000 #### 42. "text.379_of_512" 32.000000 #### 43. "text.402_of_512" 32.000000 #### 44. "text.063_of_512" 31.000000 #### 45. "text.219_of_512" 31.000000 #### 46. "text.352_of_512" 31.000000 #### 47. "text.068_of_512" 30.000000 #### 48. "text.071_of_512" 30.000000 #### 49. "text.172_of_512" 30.000000 #### 50. "text.284_of_512" 30.000000 #### 51. "text.023_of_512" 29.000000 #### 52. "text.090_of_512" 29.000000 #### 53. "text.155_of_512" 29.000000 #### 54. "text.192_of_512" 29.000000 #### 55. "text.038_of_512" 28.000000 #### 56. "text.092_of_512" 28.000000 #### 57. "text.273_of_512" 28.000000 #### 58. "text.306_of_512" 28.000000 #### 59. "text.137_of_512" 27.000000 ### 60. "text.239_of_512" 27.000000 ### 61. "text.328_of_512" 27.000000 ### 62. "text.439_of_512" 27.000000 ### 63. "text.442_of_512" 27.000000 ### 64. "text.450_of_512" 27.000000 ### 65. "text.041_of_512" 26.000000 ### 66. "text.425_of_512" 26.000000 ### 67. "text.443_of_512" 26.000000 ### 68. "text.035_of_512" 25.000000 ### 69. "text.048_of_512" 25.000000 ### 70. "text.078_of_512" 25.000000 ### 71. "text.351_of_512" 25.000000 ### 72. "text.471_of_512" 25.000000 ### 73. "text.494_of_512" 25.000000 ### 74. "text.059_of_512" 24.000000 ### 75. "text.162_of_512" 24.000000 ### 76. "text.195_of_512" 24.000000 ### 77. "text.201_of_512" 24.000000 ### 78. "text.212_of_512" 24.000000 ### 79. "text.313_of_512" 24.000000 ### 80. "text.361_of_512" 24.000000 ### 81. "text.373_of_512" 24.000000 ### 82. "text.400_of_512" 24.000000 ### 83. "text.407_of_512" 24.000000 ### 84. "text.467_of_512" 24.000000 ### 85. "text.099_of_512" 23.000000 ### 86. "text.142_of_512" 23.000000 ### 87. "text.157_of_512" 23.000000 ### 88. "text.188_of_512" 23.000000 ### 89. "text.269_of_512" 23.000000 ### 90. "text.297_of_512" 23.000000 ### 91. "text.365_of_512" 23.000000 ### 92. "text.449_of_512" 23.000000 ### 93. "text.496_of_512" 23.000000 ### 94. "text.044_of_512" 22.000000 ### 95. "text.089_of_512" 22.000000 ### 96. "text.149_of_512" 22.000000 ### 97. "text.150_of_512" 22.000000 ### 98. "text.199_of_512" 22.000000 ### 99. "text.200_of_512" 22.000000 ### 100. "text.317_of_512" 22.000000 ### 101. "text.375_of_512" 22.000000 ### 102. "text.381_of_512" 22.000000 ### 103. "text.017_of_512" 21.000000 ## 104. "text.051_of_512" 21.000000 ## 105. "text.079_of_512" 21.000000 ## 106. "text.196_of_512" 21.000000 ## 107. "text.331_of_512" 21.000000 ## 108. "text.344_of_512" 21.000000 ## 109. "text.346_of_512" 21.000000 ## 110. "text.354_of_512" 21.000000 ## 111. "text.364_of_512" 21.000000 ## 112. "text.367_of_512" 21.000000 ## 113. "text.410_of_512" 21.000000 ## 114. "text.434_of_512" 21.000000 ## 115. "text.031_of_512" 20.000000 ## 116. "text.037_of_512" 20.000000 ## 117. "text.167_of_512" 20.000000 ## 118. "text.202_of_512" 20.000000 ## 119. "text.205_of_512" 20.000000 ## 120. "text.232_of_512" 20.000000 ## 121. "text.246_of_512" 20.000000 ## 122. "text.376_of_512" 20.000000 ## 123. "text.500_of_512" 20.000000 ## 124. "text.504_of_512" 20.000000 ## 125. "text.118_of_512" 19.000000 ## 126. "text.122_of_512" 19.000000 ## 127. "text.236_of_512" 19.000000 ## 128. "text.259_of_512" 19.000000 ## 129. "text.296_of_512" 19.000000 ## 130. "text.319_of_512" 19.000000 ## 131. "text.349_of_512" 19.000000 ## 132. "text.392_of_512" 19.000000 ## 133. "text.398_of_512" 19.000000 ## 134. "text.454_of_512" 19.000000 ## 135. "text.485_of_512" 19.000000 ## 136. "text.034_of_512" 18.000000 ## 137. "text.049_of_512" 18.000000 ## 138. "text.101_of_512" 18.000000 ## 139. "text.120_of_512" 18.000000 ## 140. "text.133_of_512" 18.000000 ## 141. "text.179_of_512" 18.000000 ## 142. "text.183_of_512" 18.000000 ## 143. "text.229_of_512" 18.000000 ## 144. "text.261_of_512" 18.000000 ## 145. "text.294_of_512" 18.000000 ## 146. "text.324_of_512" 18.000000 ## 147. "text.382_of_512" 18.000000 ## 148. "text.403_of_512" 18.000000 ## 149. "text.424_of_512" 18.000000 ## 150. "text.005_of_512" 17.000000 ## 151. "text.039_of_512" 17.000000 ## 152. "text.064_of_512" 17.000000 ## 153. "text.067_of_512" 17.000000 ## 154. "text.072_of_512" 17.000000 ## 155. "text.091_of_512" 17.000000 ## 156. "text.123_of_512" 17.000000 ## 157. "text.144_of_512" 17.000000 ## 158. "text.244_of_512" 17.000000 ## 159. "text.272_of_512" 17.000000 ## 160. "text.322_of_512" 17.000000 ## 161. "text.360_of_512" 17.000000 ## 162. "text.431_of_512" 17.000000 ## 163. "text.444_of_512" 17.000000 ## 164. "text.499_of_512" 17.000000 ## 165. "text.012_of_512" 16.000000 ## 166. "text.024_of_512" 16.000000 ## 167. "text.028_of_512" 16.000000 ## 168. "text.073_of_512" 16.000000 ## 169. "text.085_of_512" 16.000000 ## 170. "text.086_of_512" 16.000000 ## 171. "text.093_of_512" 16.000000 ## 172. "text.136_of_512" 16.000000 ## 173. "text.176_of_512" 16.000000 ## 174. "text.181_of_512" 16.000000 ## 175. "text.217_of_512" 16.000000 ## 176. "text.274_of_512" 16.000000 ## 177. "text.348_of_512" 16.000000 ## 178. "text.369_of_512" 16.000000 ## 179. "text.388_of_512" 16.000000 ## 180. "text.393_of_512" 16.000000 ## 181. "text.415_of_512" 16.000000 ## 182. "text.417_of_512" 16.000000 ## 183. "text.428_of_512" 16.000000 ## 184. "text.502_of_512" 16.000000 ## 185. "text.505_of_512" 16.000000 ## 186. "text.001_of_512" 15.000000 ## 187. "text.014_of_512" 15.000000 ## 188. "text.026_of_512" 15.000000 ## 189. "text.027_of_512" 15.000000 ## 190. "text.069_of_512" 15.000000 ## 191. "text.076_of_512" 15.000000 ## 192. "text.097_of_512" 15.000000 ## 193. "text.104_of_512" 15.000000 ## 194. "text.113_of_512" 15.000000 ## 195. "text.119_of_512" 15.000000 ## 196. "text.134_of_512" 15.000000 ## 197. "text.135_of_512" 15.000000 ## 198. "text.154_of_512" 15.000000 ## 199. "text.158_of_512" 15.000000 ## 200. "text.186_of_512" 15.000000 ## 201. "text.224_of_512" 15.000000 ## 202. "text.267_of_512" 15.000000 ## 203. "text.282_of_512" 15.000000 ## 204. "text.314_of_512" 15.000000 ## 205. "text.357_of_512" 15.000000 ## 206. "text.363_of_512" 15.000000 ## 207. "text.389_of_512" 15.000000 ## 208. "text.405_of_512" 15.000000 ## 209. "text.413_of_512" 15.000000 ## 210. "text.420_of_512" 15.000000 ## 211. "text.451_of_512" 15.000000 ## 212. "text.009_of_512" 14.000000 # 213. "text.019_of_512" 14.000000 # 214. "text.025_of_512" 14.000000 # 215. "text.030_of_512" 14.000000 # 216. "text.032_of_512" 14.000000 # 217. "text.047_of_512" 14.000000 # 218. "text.054_of_512" 14.000000 # 219. "text.057_of_512" 14.000000 # 220. "text.108_of_512" 14.000000 # 221. "text.110_of_512" 14.000000 # 222. "text.114_of_512" 14.000000 # 223. "text.124_of_512" 14.000000 # 224. "text.126_of_512" 14.000000 # 225. "text.165_of_512" 14.000000 # 226. "text.208_of_512" 14.000000 # 227. "text.227_of_512" 14.000000 # 228. "text.249_of_512" 14.000000 # 229. "text.263_of_512" 14.000000 # 230. "text.265_of_512" 14.000000 # 231. "text.303_of_512" 14.000000 # 232. "text.370_of_512" 14.000000 # 233. "text.412_of_512" 14.000000 # 234. "text.448_of_512" 14.000000 # 235. "text.472_of_512" 14.000000 # 236. "text.497_of_512" 14.000000 # 237. "text.509_of_512" 14.000000 # 238. "text.002_of_512" 13.000000 # 239. "text.008_of_512" 13.000000 # 240. "text.013_of_512" 13.000000 # 241. "text.033_of_512" 13.000000 # 242. "text.066_of_512" 13.000000 # 243. "text.070_of_512" 13.000000 # 244. "text.074_of_512" 13.000000 # 245. "text.077_of_512" 13.000000 # 246. "text.100_of_512" 13.000000 # 247. "text.148_of_512" 13.000000 # 248. "text.151_of_512" 13.000000 # 249. "text.161_of_512" 13.000000 # 250. "text.184_of_512" 13.000000 # 251. "text.238_of_512" 13.000000 # 252. "text.247_of_512" 13.000000 # 253. "text.280_of_512" 13.000000 # 254. "text.292_of_512" 13.000000 # 255. "text.293_of_512" 13.000000 # 256. "text.295_of_512" 13.000000 # 257. "text.311_of_512" 13.000000 # 258. "text.337_of_512" 13.000000 # 259. "text.395_of_512" 13.000000 # 260. "text.429_of_512" 13.000000 # 261. "text.456_of_512" 13.000000 # 262. "text.458_of_512" 13.000000 # 263. "text.475_of_512" 13.000000 # 264. "text.476_of_512" 13.000000 # 265. "text.484_of_512" 13.000000 # 266. "text.003_of_512" 12.000000 # 267. "text.004_of_512" 12.000000 # 268. "text.015_of_512" 12.000000 # 269. "text.058_of_512" 12.000000 # 270. "text.061_of_512" 12.000000 # 271. "text.083_of_512" 12.000000 # 272. "text.087_of_512" 12.000000 # 273. "text.106_of_512" 12.000000 # 274. "text.109_of_512" 12.000000 # 275. "text.115_of_512" 12.000000 # 276. "text.125_of_512" 12.000000 # 277. "text.166_of_512" 12.000000 # 278. "text.175_of_512" 12.000000 # 279. "text.187_of_512" 12.000000 # 280. "text.204_of_512" 12.000000 # 281. "text.228_of_512" 12.000000 # 282. "text.241_of_512" 12.000000 # 283. "text.258_of_512" 12.000000 # 284. "text.260_of_512" 12.000000 # 285. "text.321_of_512" 12.000000 # 286. "text.339_of_512" 12.000000 # 287. "text.396_of_512" 12.000000 # 288. "text.401_of_512" 12.000000 # 289. "text.416_of_512" 12.000000 # 290. "text.426_of_512" 12.000000 # 291. "text.436_of_512" 12.000000 # 292. "text.466_of_512" 12.000000 # 293. "text.468_of_512" 12.000000 # 294. "text.469_of_512" 12.000000 # 295. "text.474_of_512" 12.000000 # 296. "text.478_of_512" 12.000000 # 297. "text.489_of_512" 12.000000 # 298. "text.492_of_512" 12.000000 # 299. "text.011_of_512" 11.000000 # 300. "text.055_of_512" 11.000000 # 301. "text.095_of_512" 11.000000 # 302. "text.105_of_512" 11.000000 # 303. "text.117_of_512" 11.000000 # 304. "text.138_of_512" 11.000000 # 305. "text.163_of_512" 11.000000 # 306. "text.169_of_512" 11.000000 # 307. "text.174_of_512" 11.000000 # 308. "text.218_of_512" 11.000000 # 309. "text.226_of_512" 11.000000 # 310. "text.275_of_512" 11.000000 # 311. "text.283_of_512" 11.000000 # 312. "text.289_of_512" 11.000000 # 313. "text.310_of_512" 11.000000 # 314. "text.315_of_512" 11.000000 # 315. "text.316_of_512" 11.000000 # 316. "text.343_of_512" 11.000000 # 317. "text.347_of_512" 11.000000 # 318. "text.372_of_512" 11.000000 # 319. "text.404_of_512" 11.000000 # 320. "text.461_of_512" 11.000000 # 321. "text.462_of_512" 11.000000 # 322. "text.464_of_512" 11.000000 # 323. "text.473_of_512" 11.000000 # 324. "text.483_of_512" 11.000000 # 325. "text.508_of_512" 11.000000 # 326. "text.102_of_512" 10.000000 # 327. "text.121_of_512" 10.000000 # 328. "text.131_of_512" 10.000000 # 329. "text.197_of_512" 10.000000 # 330. "text.209_of_512" 10.000000 # 331. "text.210_of_512" 10.000000 # 332. "text.220_of_512" 10.000000 # 333. "text.221_of_512" 10.000000 # 334. "text.225_of_512" 10.000000 # 335. "text.248_of_512" 10.000000 # 336. "text.252_of_512" 10.000000 # 337. "text.253_of_512" 10.000000 # 338. "text.256_of_512" 10.000000 # 339. "text.301_of_512" 10.000000 # 340. "text.312_of_512" 10.000000 # 341. "text.326_of_512" 10.000000 # 342. "text.330_of_512" 10.000000 # 343. "text.341_of_512" 10.000000 # 344. "text.355_of_512" 10.000000 # 345. "text.371_of_512" 10.000000 # 346. "text.391_of_512" 10.000000 # 347. "text.421_of_512" 10.000000 # 348. "text.430_of_512" 10.000000 # 349. "text.440_of_512" 10.000000 # 350. "text.445_of_512" 10.000000 # 351. "text.446_of_512" 10.000000 # 352. "text.465_of_512" 10.000000 # 353. "text.481_of_512" 10.000000 # 354. "text.486_of_512" 10.000000 # 355. "text.000_of_512" 9.000000 # 356. "text.018_of_512" 9.000000 # 357. "text.084_of_512" 9.000000 # 358. "text.116_of_512" 9.000000 # 359. "text.143_of_512" 9.000000 # 360. "text.159_of_512" 9.000000 # 361. "text.164_of_512" 9.000000 # 362. "text.178_of_512" 9.000000 # 363. "text.189_of_512" 9.000000 # 364. "text.191_of_512" 9.000000 # 365. "text.213_of_512" 9.000000 # 366. "text.215_of_512" 9.000000 # 367. "text.223_of_512" 9.000000 # 368. "text.243_of_512" 9.000000 # 369. "text.264_of_512" 9.000000 # 370. "text.266_of_512" 9.000000 # 371. "text.291_of_512" 9.000000 # 372. "text.304_of_512" 9.000000 # 373. "text.325_of_512" 9.000000 # 374. "text.327_of_512" 9.000000 # 375. "text.335_of_512" 9.000000 # 376. "text.336_of_512" 9.000000 # 377. "text.374_of_512" 9.000000 # 378. "text.387_of_512" 9.000000 # 379. "text.390_of_512" 9.000000 # 380. "text.397_of_512" 9.000000 # 381. "text.418_of_512" 9.000000 # 382. "text.433_of_512" 9.000000 # 383. "text.479_of_512" 9.000000 # 384. "text.480_of_512" 9.000000 # 385. "text.487_of_512" 9.000000 # 386. "text.493_of_512" 9.000000 # 387. "text.495_of_512" 9.000000 # 388. "text.007_of_512" 8.000000 389. "text.010_of_512" 8.000000 390. "text.016_of_512" 8.000000 391. "text.020_of_512" 8.000000 392. "text.036_of_512" 8.000000 393. "text.040_of_512" 8.000000 394. "text.042_of_512" 8.000000 395. "text.052_of_512" 8.000000 396. "text.053_of_512" 8.000000 397. "text.062_of_512" 8.000000 398. "text.075_of_512" 8.000000 399. "text.128_of_512" 8.000000 400. "text.130_of_512" 8.000000 401. "text.145_of_512" 8.000000 402. "text.147_of_512" 8.000000 403. "text.170_of_512" 8.000000 404. "text.177_of_512" 8.000000 405. "text.203_of_512" 8.000000 406. "text.207_of_512" 8.000000 407. "text.211_of_512" 8.000000 408. "text.214_of_512" 8.000000 409. "text.216_of_512" 8.000000 410. "text.222_of_512" 8.000000 411. "text.254_of_512" 8.000000 412. "text.255_of_512" 8.000000 413. "text.277_of_512" 8.000000 414. "text.279_of_512" 8.000000 415. "text.286_of_512" 8.000000 416. "text.333_of_512" 8.000000 417. "text.345_of_512" 8.000000 418. "text.358_of_512" 8.000000 419. "text.359_of_512" 8.000000 420. "text.366_of_512" 8.000000 421. "text.422_of_512" 8.000000 422. "text.423_of_512" 8.000000 423. "text.447_of_512" 8.000000 424. "text.457_of_512" 8.000000 425. "text.498_of_512" 8.000000 426. "text.506_of_512" 8.000000 427. "text.021_of_512" 7.000000 428. "text.050_of_512" 7.000000 429. "text.080_of_512" 7.000000 430. "text.094_of_512" 7.000000 431. "text.096_of_512" 7.000000 432. "text.107_of_512" 7.000000 433. "text.112_of_512" 7.000000 434. "text.141_of_512" 7.000000 435. "text.146_of_512" 7.000000 436. "text.156_of_512" 7.000000 437. "text.160_of_512" 7.000000 438. "text.180_of_512" 7.000000 439. "text.185_of_512" 7.000000 440. "text.230_of_512" 7.000000 441. "text.233_of_512" 7.000000 442. "text.234_of_512" 7.000000 443. "text.287_of_512" 7.000000 444. "text.320_of_512" 7.000000 445. "text.323_of_512" 7.000000 446. "text.329_of_512" 7.000000 447. "text.332_of_512" 7.000000 448. "text.368_of_512" 7.000000 449. "text.409_of_512" 7.000000 450. "text.411_of_512" 7.000000 451. "text.432_of_512" 7.000000 452. "text.438_of_512" 7.000000 453. "text.441_of_512" 7.000000 454. "text.460_of_512" 7.000000 455. "text.488_of_512" 7.000000 456. "text.511_of_512" 7.000000 457. "text.029_of_512" 6.000000 458. "text.060_of_512" 6.000000 459. "text.081_of_512" 6.000000 460. "text.139_of_512" 6.000000 461. "text.193_of_512" 6.000000 462. "text.206_of_512" 6.000000 463. "text.231_of_512" 6.000000 464. "text.235_of_512" 6.000000 465. "text.268_of_512" 6.000000 466. "text.299_of_512" 6.000000 467. "text.338_of_512" 6.000000 468. "text.362_of_512" 6.000000 469. "text.378_of_512" 6.000000 470. "text.380_of_512" 6.000000 471. "text.394_of_512" 6.000000 472. "text.490_of_512" 6.000000 473. "text.491_of_512" 6.000000 474. "text.501_of_512" 6.000000 475. "text.503_of_512" 6.000000 476. "text.507_of_512" 6.000000 477. "text.088_of_512" 5.000000 478. "text.103_of_512" 5.000000 479. "text.129_of_512" 5.000000 480. "text.152_of_512" 5.000000 481. "text.242_of_512" 5.000000 482. "text.271_of_512" 5.000000 483. "text.298_of_512" 5.000000 484. "text.309_of_512" 5.000000 485. "text.334_of_512" 5.000000 486. "text.356_of_512" 5.000000 487. "text.377_of_512" 5.000000 488. "text.385_of_512" 5.000000 489. "text.437_of_512" 5.000000 490. "text.470_of_512" 5.000000 491. "text.477_of_512" 5.000000 492. "text.127_of_512" 4.000000 493. "text.153_of_512" 4.000000 494. "text.257_of_512" 4.000000 495. "text.278_of_512" 4.000000 496. "text.290_of_512" 4.000000 497. "text.340_of_512" 4.000000 498. "text.406_of_512" 4.000000 499. "text.414_of_512" 4.000000 500. "text.453_of_512" 4.000000 501. "text.098_of_512" 3.000000 502. "text.262_of_512" 3.000000 503. "text.318_of_512" 3.000000 504. "text.383_of_512" 3.000000 505. "text.427_of_512" 3.000000 506. "text.455_of_512" 3.000000 507. "text.250_of_512" 2.000000 508. "text.270_of_512" 2.000000 509. "text.285_of_512" 2.000000 510. "text.288_of_512" 2.000000 511. "text.399_of_512" 2.000000
1. "text.111_of_512" 5317.440705 ################ 2. "text.171_of_512" 3650.971216 ########## 3. "text.022_of_512" 2107.119045 ###### 4. "text.408_of_512" 1258.476515 ### 5. "text.384_of_512" 583.386139 # 6. "text.182_of_512" 521.017198 # 7. "text.245_of_512" 497.013337 # 8. "text.132_of_512" 440.421780 # 9. "text.065_of_512" 288.803408 10. "text.308_of_512" 260.244628 11. "text.240_of_512" 186.417431 12. "text.452_of_512" 178.882914 13. "text.302_of_512" 171.531311 14. "text.386_of_512" 165.186666 15. "text.194_of_512" 148.646312 16. "text.046_of_512" 146.422312 17. "text.281_of_512" 141.637157 18. "text.276_of_512" 140.359151 19. "text.068_of_512" 129.013294 20. "text.251_of_512" 122.035067 21. "text.082_of_512" 120.480551 22. "text.342_of_512" 119.241819 23. "text.155_of_512" 116.133530 24. "text.006_of_512" 114.005498 25. "text.237_of_512" 107.948035 26. "text.198_of_512" 102.915797 27. "text.307_of_512" 101.383260 28. "text.459_of_512" 100.613554 29. "text.305_of_512" 99.750798 30. "text.190_of_512" 97.779025 31. "text.168_of_512" 86.237237 32. "text.410_of_512" 81.357461 33. "text.300_of_512" 76.465978 34. "text.350_of_512" 75.684146 35. "text.373_of_512" 71.727791 36. "text.419_of_512" 70.934259 37. "text.045_of_512" 70.750398 38. "text.494_of_512" 68.736289 39. "text.273_of_512" 67.399279 40. "text.092_of_512" 67.257275 41. "text.463_of_512" 65.708981 42. "text.443_of_512" 63.838852 43. "text.056_of_512" 63.255381 44. "text.041_of_512" 62.847803 45. "text.063_of_512" 61.377023 46. "text.172_of_512" 60.308579 47. "text.219_of_512" 57.861603 48. "text.400_of_512" 57.415557 49. "text.239_of_512" 57.209379 50. "text.038_of_512" 56.844985 51. "text.099_of_512" 55.897056 52. "text.023_of_512" 55.701883 53. "text.188_of_512" 51.628168 54. "text.173_of_512" 51.329455 55. "text.142_of_512" 50.750685 56. "text.352_of_512" 50.733107 57. "text.435_of_512" 50.508978 58. "text.140_of_512" 50.480688 59. "text.354_of_512" 50.449122 60. "text.192_of_512" 49.547294 61. "text.035_of_512" 48.621379 62. "text.482_of_512" 47.790082 63. "text.402_of_512" 46.763297 64. "text.246_of_512" 44.509470 65. "text.375_of_512" 43.320333 66. "text.071_of_512" 42.733239 67. "text.137_of_512" 42.413625 68. "text.407_of_512" 42.250061 69. "text.090_of_512" 39.034988 70. "text.079_of_512" 38.931682 71. "text.379_of_512" 38.113812 72. "text.284_of_512" 36.703088 73. "text.328_of_512" 35.433745 74. "text.425_of_512" 35.315613 75. "text.205_of_512" 35.210872 76. "text.361_of_512" 35.152419 77. "text.059_of_512" 35.057900 78. "text.043_of_512" 34.466540 79. "text.319_of_512" 34.310743 80. "text.353_of_512" 33.269670 81. "text.091_of_512" 33.242414 82. "text.442_of_512" 32.977177 83. "text.504_of_512" 32.960466 84. "text.051_of_512" 32.868470 85. "text.367_of_512" 32.653329 86. "text.195_of_512" 32.347286 87. "text.291_of_512" 31.768630 88. "text.450_of_512" 31.272031 89. "text.502_of_512" 30.745872 90. "text.150_of_512" 30.198868 91. "text.365_of_512" 29.514659 92. "text.364_of_512" 28.440199 93. "text.464_of_512" 28.430887 94. "text.471_of_512" 27.941904 95. "text.439_of_512" 27.632742 96. "text.496_of_512" 26.517321 97. "text.201_of_512" 26.215863 98. "text.202_of_512" 26.088579 99. "text.089_of_512" 25.860687 100. "text.313_of_512" 25.797199 101. "text.078_of_512" 25.426369 102. "text.212_of_512" 25.281870 103. "text.485_of_512" 24.886574 104. "text.339_of_512" 24.763237 105. "text.269_of_512" 24.235255 106. "text.381_of_512" 24.130311 107. "text.181_of_512" 23.897232 108. "text.179_of_512" 23.621462 109. "text.118_of_512" 23.510809 110. "text.376_of_512" 23.438191 111. "text.048_of_512" 23.408248 112. "text.069_of_512" 23.167154 113. "text.349_of_512" 22.549046 114. "text.306_of_512" 22.461515 115. "text.126_of_512" 22.373418 116. "text.034_of_512" 22.286870 117. "text.009_of_512" 22.139563 118. "text.017_of_512" 22.114909 119. "text.157_of_512" 22.023544 120. "text.167_of_512" 21.659745 121. "text.064_of_512" 21.529417 122. "text.331_of_512" 21.519391 123. "text.454_of_512" 21.366435 124. "text.120_of_512" 21.327486 125. "text.200_of_512" 21.311714 126. "text.403_of_512" 21.297439 127. "text.297_of_512" 21.296606 128. "text.093_of_512" 21.235822 129. "text.259_of_512" 21.174789 130. "text.346_of_512" 21.104595 131. "text.369_of_512" 21.071965 132. "text.176_of_512" 21.052389 133. "text.282_of_512" 20.987490 134. "text.232_of_512" 20.505988 135. "text.424_of_512" 20.001049 136. "text.500_of_512" 19.965351 137. "text.224_of_512" 19.929336 138. "text.317_of_512" 19.924588 139. "text.467_of_512" 19.778128 140. "text.026_of_512" 19.717116 141. "text.097_of_512" 19.493523 142. "text.429_of_512" 19.442672 143. "text.100_of_512" 19.348618 144. "text.178_of_512" 19.139152 145. "text.388_of_512" 19.045348 146. "text.196_of_512" 18.909742 147. "text.327_of_512" 18.793089 148. "text.434_of_512" 18.471805 149. "text.227_of_512" 18.254487 150. "text.351_of_512" 18.220191 151. "text.134_of_512" 18.173989 152. "text.382_of_512" 17.996724 153. "text.049_of_512" 17.900983 154. "text.044_of_512" 17.888628 155. "text.067_of_512" 17.879445 156. "text.322_of_512" 17.821064 157. "text.247_of_512" 17.748073 158. "text.420_of_512" 17.736192 159. "text.274_of_512" 17.718408 160. "text.314_of_512" 17.661533 161. "text.149_of_512" 17.626049 162. "text.028_of_512" 17.605569 163. "text.031_of_512" 17.462220 164. "text.162_of_512" 17.402495 165. "text.229_of_512" 17.355854 166. "text.039_of_512" 17.155819 167. "text.324_of_512" 17.068312 168. "text.263_of_512" 16.989021 169. "text.199_of_512" 16.823877 170. "text.108_of_512" 16.818034 171. "text.396_of_512" 16.750610 172. "text.085_of_512" 16.664209 173. "text.499_of_512" 16.548145 174. "text.077_of_512" 16.506279 175. "text.037_of_512" 16.441114 176. "text.217_of_512" 16.201147 177. "text.479_of_512" 16.087491 178. "text.347_of_512" 16.014063 179. "text.296_of_512" 15.917289 180. "text.159_of_512" 15.869470 181. "text.123_of_512" 15.811632 182. "text.449_of_512" 15.754831 183. "text.184_of_512" 15.562026 184. "text.073_of_512" 15.511483 185. "text.294_of_512" 15.496874 186. "text.086_of_512" 15.474732 187. "text.122_of_512" 15.464668 188. "text.183_of_512" 15.443564 189. "text.363_of_512" 15.438860 190. "text.193_of_512" 15.386683 191. "text.236_of_512" 15.318106 192. "text.165_of_512" 15.231833 193. "text.505_of_512" 15.205128 194. "text.295_of_512" 15.182844 195. "text.144_of_512" 15.135931 196. "text.272_of_512" 15.063292 197. "text.344_of_512" 14.938082 198. "text.101_of_512" 14.927759 199. "text.249_of_512" 14.773932 200. "text.187_of_512" 14.773709 201. "text.072_of_512" 14.645675 202. "text.027_of_512" 14.611296 203. "text.444_of_512" 14.502462 204. "text.497_of_512" 14.360451 205. "text.104_of_512" 14.193867 206. "text.484_of_512" 14.094463 207. "text.428_of_512" 14.073558 208. "text.311_of_512" 14.027835 209. "text.047_of_512" 13.774420 210. "text.392_of_512" 13.729674 211. "text.451_of_512" 13.647917 212. "text.478_of_512" 13.525068 213. "text.509_of_512" 13.460942 214. "text.208_of_512" 13.436668 215. "text.024_of_512" 13.425313 216. "text.109_of_512" 13.351326 217. "text.431_of_512" 13.281562 218. "text.415_of_512" 13.257779 219. "text.348_of_512" 13.200316 220. "text.289_of_512" 13.189908 221. "text.261_of_512" 13.179039 222. "text.238_of_512" 13.125872 223. "text.389_of_512" 13.121540 224. "text.357_of_512" 13.084106 225. "text.398_of_512" 13.014128 226. "text.417_of_512" 12.894288 227. "text.013_of_512" 12.791177 228. "text.360_of_512" 12.687461 229. "text.158_of_512" 12.680247 230. "text.106_of_512" 12.614829 231. "text.014_of_512" 12.607888 232. "text.405_of_512" 12.540806 233. "text.054_of_512" 12.533079 234. "text.105_of_512" 12.459868 235. "text.053_of_512" 12.449502 236. "text.391_of_512" 12.403246 237. "text.330_of_512" 12.388499 238. "text.076_of_512" 12.320446 239. "text.110_of_512" 12.267189 240. "text.133_of_512" 12.177895 241. "text.292_of_512" 12.125224 242. "text.244_of_512" 12.117297 243. "text.226_of_512" 12.055183 244. "text.114_of_512" 11.989928 245. "text.487_of_512" 11.967709 246. "text.087_of_512" 11.840567 247. "text.015_of_512" 11.827488 248. "text.151_of_512" 11.792439 249. "text.154_of_512" 11.725545 250. "text.241_of_512" 11.695179 251. "text.121_of_512" 11.677985 252. "text.404_of_512" 11.672857 253. "text.012_of_512" 11.653154 254. "text.002_of_512" 11.547923 255. "text.030_of_512" 11.524788 256. "text.135_of_512" 11.430597 257. "text.057_of_512" 11.369234 258. "text.267_of_512" 11.339672 259. "text.175_of_512" 11.319826 260. "text.413_of_512" 11.240387 261. "text.321_of_512" 11.224585 262. "text.412_of_512" 11.189550 263. "text.161_of_512" 11.162601 264. "text.055_of_512" 11.074766 265. "text.370_of_512" 11.044554 266. "text.020_of_512" 11.025795 267. "text.148_of_512" 10.991195 268. "text.343_of_512" 10.978523 269. "text.395_of_512" 10.940141 270. "text.436_of_512" 10.924356 271. "text.125_of_512" 10.913189 272. "text.066_of_512" 10.893415 273. "text.508_of_512" 10.880481 274. "text.042_of_512" 10.755142 275. "text.220_of_512" 10.738125 276. "text.416_of_512" 10.661602 277. "text.456_of_512" 10.661070 278. "text.005_of_512" 10.502730 279. "text.458_of_512" 10.490514 280. "text.204_of_512" 10.458747 281. "text.019_of_512" 10.450397 282. "text.372_of_512" 10.400262 283. "text.472_of_512" 10.383116 284. "text.441_of_512" 10.317965 285. "text.326_of_512" 10.285882 286. "text.258_of_512" 10.265295 287. "text.506_of_512" 10.250057 288. "text.115_of_512" 10.241683 289. "text.136_of_512" 10.216008 290. "text.225_of_512" 10.182855 291. "text.117_of_512" 10.166464 292. "text.228_of_512" 10.106576 293. "text.492_of_512" 10.066116 294. "text.474_of_512" 10.057981 295. "text.032_of_512" 10.040702 296. "text.248_of_512" 9.938959 297. "text.260_of_512" 9.886765 298. "text.021_of_512" 9.867193 299. "text.401_of_512" 9.828373 300. "text.095_of_512" 9.769616 301. "text.483_of_512" 9.729457 302. "text.050_of_512" 9.706603 303. "text.475_of_512" 9.611140 304. "text.303_of_512" 9.605214 305. "text.211_of_512" 9.538452 306. "text.070_of_512" 9.529355 307. "text.337_of_512" 9.424925 308. "text.124_of_512" 9.424201 309. "text.230_of_512" 9.359748 310. "text.433_of_512" 9.354902 311. "text.119_of_512" 9.319691 312. "text.448_of_512" 9.272183 313. "text.001_of_512" 9.259415 314. "text.210_of_512" 9.249007 315. "text.138_of_512" 9.219237 316. "text.058_of_512" 9.131830 317. "text.469_of_512" 9.125371 318. "text.008_of_512" 9.124477 319. "text.275_of_512" 9.112806 320. "text.083_of_512" 9.050527 321. "text.316_of_512" 8.956921 322. "text.025_of_512" 8.925180 323. "text.476_of_512" 8.902608 324. "text.312_of_512" 8.896903 325. "text.004_of_512" 8.882758 326. "text.280_of_512" 8.829254 327. "text.465_of_512" 8.823126 328. "text.156_of_512" 8.812926 329. "text.461_of_512" 8.747426 330. "text.003_of_512" 8.713203 331. "text.113_of_512" 8.685859 332. "text.393_of_512" 8.644423 333. "text.265_of_512" 8.589267 334. "text.256_of_512" 8.565640 335. "text.495_of_512" 8.511515 336. "text.255_of_512" 8.503298 337. "text.145_of_512" 8.479545 338. "text.293_of_512" 8.454661 339. "text.387_of_512" 8.444664 340. "text.423_of_512" 8.425266 341. "text.166_of_512" 8.422371 342. "text.298_of_512" 8.418135 343. "text.466_of_512" 8.392731 344. "text.253_of_512" 8.387091 345. "text.468_of_512" 8.378500 346. "text.084_of_512" 8.373563 347. "text.223_of_512" 8.300315 348. "text.033_of_512" 8.290740 349. "text.481_of_512" 8.199431 350. "text.186_of_512" 8.167259 351. "text.169_of_512" 8.043737 352. "text.215_of_512" 7.998535 353. "text.341_of_512" 7.990455 354. "text.213_of_512" 7.947965 355. "text.422_of_512" 7.883293 356. "text.000_of_512" 7.873737 357. "text.036_of_512" 7.833242 358. "text.061_of_512" 7.794338 359. "text.304_of_512" 7.791422 360. "text.440_of_512" 7.772200 361. "text.074_of_512" 7.720522 362. "text.325_of_512" 7.701480 363. "text.174_of_512" 7.648511 364. "text.473_of_512" 7.640396 365. "text.116_of_512" 7.432516 366. "text.088_of_512" 7.375702 367. "text.462_of_512" 7.370414 368. "text.141_of_512" 7.358679 369. "text.315_of_512" 7.292697 370. "text.320_of_512" 7.285890 371. "text.052_of_512" 7.263827 372. "text.371_of_512" 7.211039 373. "text.301_of_512" 7.204238 374. "text.390_of_512" 7.184831 375. "text.214_of_512" 7.165112 376. "text.310_of_512" 7.120250 377. "text.152_of_512" 7.114524 378. "text.127_of_512" 7.105350 379. "text.062_of_512" 7.039762 380. "text.143_of_512" 6.995768 381. "text.107_of_512" 6.992239 382. "text.438_of_512" 6.987579 383. "text.131_of_512" 6.976154 384. "text.432_of_512" 6.925402 385. "text.329_of_512" 6.898728 386. "text.147_of_512" 6.843268 387. "text.355_of_512" 6.840869 388. "text.406_of_512" 6.778595 389. "text.446_of_512" 6.766229 390. "text.489_of_512" 6.745740 391. "text.075_of_512" 6.739690 392. "text.218_of_512" 6.688297 393. "text.453_of_512" 6.650884 394. "text.426_of_512" 6.584627 395. "text.335_of_512" 6.569409 396. "text.358_of_512" 6.532798 397. "text.493_of_512" 6.523429 398. "text.007_of_512" 6.411961 399. "text.409_of_512" 6.410372 400. "text.460_of_512" 6.402201 401. "text.011_of_512" 6.397151 402. "text.503_of_512" 6.354757 403. "text.283_of_512" 6.328816 404. "text.366_of_512" 6.263960 405. "text.197_of_512" 6.246095 406. "text.243_of_512" 6.194773 407. "text.421_of_512" 6.105908 408. "text.374_of_512" 6.097693 409. "text.277_of_512" 6.068572 410. "text.332_of_512" 6.067873 411. "text.252_of_512" 6.036750 412. "text.345_of_512" 6.021022 413. "text.191_of_512" 5.988329 414. "text.491_of_512" 5.984841 415. "text.418_of_512" 5.967021 416. "text.203_of_512" 5.944533 417. "text.018_of_512" 5.939225 418. "text.397_of_512" 5.903827 419. "text.139_of_512" 5.829417 420. "text.163_of_512" 5.752617 421. "text.130_of_512" 5.700992 422. "text.231_of_512" 5.688939 423. "text.457_of_512" 5.672374 424. "text.153_of_512" 5.663403 425. "text.368_of_512" 5.651593 426. "text.102_of_512" 5.646376 427. "text.486_of_512" 5.633483 428. "text.216_of_512" 5.629287 429. "text.264_of_512" 5.588973 430. "text.385_of_512" 5.584986 431. "text.490_of_512" 5.562708 432. "text.411_of_512" 5.551111 433. "text.501_of_512" 5.525986 434. "text.016_of_512" 5.516382 435. "text.254_of_512" 5.513080 436. "text.209_of_512" 5.467364 437. "text.498_of_512" 5.420012 438. "text.080_of_512" 5.374728 439. "text.128_of_512" 5.347706 440. "text.359_of_512" 5.270495 441. "text.189_of_512" 5.253403 442. "text.222_of_512" 5.217164 443. "text.221_of_512" 5.178786 444. "text.362_of_512" 5.137636 445. "text.445_of_512" 5.128978 446. "text.177_of_512" 5.116982 447. "text.488_of_512" 5.095297 448. "text.334_of_512" 5.072061 449. "text.430_of_512" 5.053250 450. "text.060_of_512" 5.042195 451. "text.279_of_512" 5.035647 452. "text.378_of_512" 5.022113 453. "text.336_of_512" 4.982412 454. "text.010_of_512" 4.969288 455. "text.480_of_512" 4.966521 456. "text.164_of_512" 4.893784 457. "text.286_of_512" 4.884769 458. "text.323_of_512" 4.852872 459. "text.040_of_512" 4.820518 460. "text.081_of_512" 4.763537 461. "text.333_of_512" 4.742724 462. "text.029_of_512" 4.704571 463. "text.233_of_512" 4.688918 464. "text.207_of_512" 4.668682 465. "text.112_of_512" 4.648415 466. "text.511_of_512" 4.635224 467. "text.146_of_512" 4.626896 468. "text.338_of_512" 4.585168 469. "text.096_of_512" 4.329233 470. "text.287_of_512" 4.263560 471. "text.377_of_512" 4.234776 472. "text.266_of_512" 4.231089 473. "text.206_of_512" 4.208843 474. "text.170_of_512" 4.173307 475. "text.103_of_512" 4.137199 476. "text.094_of_512" 4.077613 477. "text.160_of_512" 4.010110 478. "text.180_of_512" 4.002525 479. "text.185_of_512" 3.965777 480. "text.290_of_512" 3.930396 481. "text.447_of_512" 3.871413 482. "text.380_of_512" 3.856945 483. "text.098_of_512" 3.817470 484. "text.394_of_512" 3.728094 485. "text.234_of_512" 3.596383 486. "text.477_of_512" 3.430310 487. "text.507_of_512" 3.401816 488. "text.278_of_512" 3.371364 489. "text.235_of_512" 3.351894 490. "text.340_of_512" 3.257311 491. "text.437_of_512" 3.248597 492. "text.268_of_512" 3.169426 493. "text.242_of_512" 3.129669 494. "text.414_of_512" 2.998142 495. "text.299_of_512" 2.978278 496. "text.129_of_512" 2.929291 497. "text.470_of_512" 2.814402 498. "text.309_of_512" 2.749645 499. "text.427_of_512" 2.535817 500. "text.318_of_512" 2.409882 501. "text.383_of_512" 2.354953 502. "text.271_of_512" 2.329468 503. "text.455_of_512" 2.312649 504. "text.257_of_512" 1.948737 505. "text.356_of_512" 1.793478 506. "text.262_of_512" 1.436583 507. "text.285_of_512" 1.325838 508. "text.250_of_512" 0.972889 509. "text.270_of_512" 0.908637 510. "text.399_of_512" 0.842241 511. "text.288_of_512" 0.562196
Those variable importances are computed during training. More, and possibly more informative, variable importances are available when analyzing a model on a test dataset.
Only printing the first tree.
Tree #0: "text.111_of_512">=-0.0602566 [s:0.0555518 n:22523 np:11909 miss:1] ; pred:5.56483e-09 ├─(pos)─ "text.171_of_512">=-0.0261862 [s:0.0197489 n:11909 np:5560 miss:0] ; pred:0.0890043 | ├─(pos)─ "text.111_of_512">=0.00956145 [s:0.00532368 n:5560 np:3094 miss:0] ; pred:0.149073 | | ├─(pos)─ "text.065_of_512">=0.0368076 [s:0.00144737 n:3094 np:2189 miss:0] ; pred:0.175128 | | | ├─(pos)─ "text.219_of_512">=-0.0505817 [s:0.00122259 n:2189 np:2087 miss:1] ; pred:0.184913 | | | | ├─(pos)─ pred:0.188005 | | | | └─(neg)─ pred:0.121649 | | | └─(neg)─ "text.302_of_512">=-0.0171091 [s:0.00536794 n:905 np:823 miss:1] ; pred:0.151461 | | | ├─(pos)─ pred:0.160712 | | | └─(neg)─ pred:0.0586165 | | └─(neg)─ "text.022_of_512">=0.0102688 [s:0.00895105 n:2466 np:1129 miss:0] ; pred:0.116382 | | ├─(pos)─ "text.408_of_512">=0.0202419 [s:0.00306305 n:1129 np:772 miss:0] ; pred:0.157564 | | | ├─(pos)─ pred:0.172619 | | | └─(neg)─ pred:0.12501 | | └─(neg)─ "text.408_of_512">=0.0430553 [s:0.0116265 n:1337 np:495 miss:0] ; pred:0.0816057 | | ├─(pos)─ pred:0.137858 | | └─(neg)─ pred:0.048536 | └─(neg)─ "text.022_of_512">=-0.0110774 [s:0.023812 n:6349 np:3846 miss:1] ; pred:0.0364006 | ├─(pos)─ "text.182_of_512">=-0.0230205 [s:0.0119394 n:3846 np:2589 miss:1] ; pred:0.0861954 | | ├─(pos)─ "text.022_of_512">=0.049516 [s:0.00617494 n:2589 np:1240 miss:0] ; pred:0.11665 | | | ├─(pos)─ pred:0.149435 | | | └─(neg)─ pred:0.0865143 | | └─(neg)─ "text.410_of_512">=0.00820597 [s:0.0146238 n:1257 np:362 miss:0] ; pred:0.023469 | | ├─(pos)─ pred:0.0995275 | | └─(neg)─ pred:-0.00729438 | └─(neg)─ "text.171_of_512">=-0.0539957 [s:0.0143023 n:2503 np:1285 miss:1] ; pred:-0.0401119 | ├─(pos)─ "text.384_of_512">=0.0559715 [s:0.0131526 n:1285 np:299 miss:0] ; pred:0.00646125 | | ├─(pos)─ pred:-0.0768432 | | └─(neg)─ pred:0.0317229 | └─(neg)─ "text.408_of_512">=0.0233407 [s:0.00972599 n:1218 np:334 miss:0] ; pred:-0.0892469 | ├─(pos)─ pred:-0.0250698 | └─(neg)─ pred:-0.113495 └─(neg)─ "text.408_of_512">=0.0146363 [s:0.0204198 n:10614 np:2509 miss:0] ; pred:-0.0998636 ├─(pos)─ "text.022_of_512">=0.0184357 [s:0.0245182 n:2509 np:1137 miss:0] ; pred:0.00286988 | ├─(pos)─ "text.452_of_512">=0.000155947 [s:0.0117618 n:1137 np:442 miss:1] ; pred:0.0716719 | | ├─(pos)─ "text.182_of_512">=0.00088577 [s:0.0162757 n:442 np:243 miss:0] ; pred:0.0172745 | | | ├─(pos)─ pred:0.0634544 | | | └─(neg)─ pred:-0.0391161 | | └─(neg)─ "text.182_of_512">=-0.0310871 [s:0.0106962 n:695 np:597 miss:1] ; pred:0.106267 | | ├─(pos)─ pred:0.123028 | | └─(neg)─ pred:0.00416156 | └─(neg)─ "text.171_of_512">=-0.0535029 [s:0.0226459 n:1372 np:394 miss:1] ; pred:-0.0541475 | ├─(pos)─ "text.291_of_512">=0.0608138 [s:0.0252759 n:394 np:318 miss:1] ; pred:0.0406891 | | ├─(pos)─ pred:0.0717781 | | └─(neg)─ pred:-0.0893938 | └─(neg)─ "text.305_of_512">=-0.00659941 [s:0.0113343 n:978 np:282 miss:0] ; pred:-0.0923536 | ├─(pos)─ pred:-0.025452 | └─(neg)─ pred:-0.11946 └─(neg)─ "text.022_of_512">=0.0319392 [s:0.010925 n:8105 np:1696 miss:0] ; pred:-0.131666 ├─(pos)─ "text.245_of_512">=0.0584148 [s:0.016193 n:1696 np:644 miss:0] ; pred:-0.0503918 | ├─(pos)─ "text.173_of_512">=0.0489011 [s:0.0113042 n:644 np:64 miss:0] ; pred:-0.115448 | | ├─(pos)─ pred:0.0125799 | | └─(neg)─ pred:-0.129575 | └─(neg)─ "text.384_of_512">=0.0196009 [s:0.0159509 n:1052 np:800 miss:1] ; pred:-0.0105665 | ├─(pos)─ pred:-0.0389201 | └─(neg)─ pred:0.079445 └─(neg)─ "text.245_of_512">=0.0363683 [s:0.00411718 n:6409 np:4305 miss:0] ; pred:-0.153173 ├─(pos)─ "text.171_of_512">=-0.0644174 [s:0.00162702 n:4305 np:1069 miss:1] ; pred:-0.171116 | ├─(pos)─ pred:-0.143045 | └─(neg)─ pred:-0.18039 └─(neg)─ "text.171_of_512">=-0.0632608 [s:0.00815962 n:2104 np:628 miss:1] ; pred:-0.11646 ├─(pos)─ pred:-0.0610666 └─(neg)─ pred:-0.140029
Evaluating model¶
We evaluate the model on the test dataset.
model.evaluate(test_ds)
Evaluation of classification models
- Accuracy
- The simplest metric. It's the percentage of predictions that are correct (matching the ground truth).
Example: If a model correctly identifies 90 out of 100 images as cat or dog, the accuracy is 90%. - Confusion Matrix
- A table that shows the counts of:
- True Positives (TP): Model correctly predicted positive.
- True Negatives (TN): Model correctly predicted negative.
- False Positives (FP): Model incorrectly predicted positive (a "false alarm").
- False Negatives (FN): Model incorrectly predicted negative (a "miss").
- Threshold
- YDF classification models predict a probability for each class. A threshold determines the cutoff for classifying something as positive or negative.
Example: If the threshold is 0.5, any prediction above 0.5 might be classified as "spam," and anything below as "not spam." - ROC Curve (Receiver Operating Characteristic Curve)
- A graph that plots the True Positive Rate (TPR) against the False Positive Rate (FPR) at various thresholds.
- TPR (Sensitivity or Recall): TP / (TP + FN) - How many of the actual positives did the model catch?
- FPR: FP / (FP + TN) - How many negatives were incorrectly classified as positives?
Interpretation: A good model has an ROC curve that hugs the top-left corner (high TPR, low FPR). - AUC (Area Under the ROC Curve)
- A single number that summarizes the overall performance shown by the ROC curve. The AUC is a more stable metric than the accuracy. Multi-class classification models evaluate one class against all other classes.
Interpretation: Ranges from 0 to 1. A perfect model has an AUC of 1, while a random model has an AUC of 0.5. Higher is better. - Precision-Recall Curve
- A graph that plots Precision against Recall at various thresholds.
- Precision: TP / (TP + FP) - Out of all the predictions the model labeled as positive, how many were actually positive?
- Recall (same as TPR): TP / (TP + FN) - Out of all the actual positive cases, how many did the model correctly identify?
Interpretation: A good model has a curve that stays high (both high precision and high recall). It is especially useful when dealing with imbalanced datasets (e.g., when one class is much rarer than the other). - PR-AUC (Area Under the Precision-Recall Curve)
- Similar to AUC, but for the Precision-Recall curve. A single number summarizing performance. Multi-class classification models evaluate one class against all other classes. Higher is better.
- Threshold / Accuracy Curve
- A graph that shows how the model's accuracy changes as you vary the classification threshold.
- Threshold / Volume Curve
- A graph showing how the number of data points classified as positive changes as you vary the threshold.
Label \ Pred | 0 | 1 |
---|---|---|
0 | 10724 | 1941 |
1 | 1776 | 10559 |