search

google-research / tapas

End-to-end neural table-text understanding models.
Apache License 2.0
1.15k stars 217 forks source link

Questions about predict on the TABFACT task #97

Closed wcc17864158993 closed 3 years ago

wcc17864158993 commented 3 years ago

Hi, I have found a question when I was running the codes with "predict" mode.

If I used the following codes:

import tensorflow.compat.v1 as tf
import os 
import shutil
import csv
import pandas as pd
import IPython

tf.get_logger().setLevel('ERROR')

from tapas.utils import tf_example_utils
from tapas.protos import interaction_pb2
from tapas.utils import number_annotation_utils
import math
import json

os.makedirs('predict/tabfact/tf_examples', exist_ok=True)
os.makedirs('predict/tabfact/model', exist_ok=True)
with open('predict/tabfact/model/checkpoint', 'w') as f:
    f.write('model_checkpoint_path: "model.ckpt-0"')
for suffix in ['model.ckpt-8000.data-00000-of-00001', 'model.ckpt-8000.index', 'model.ckpt-8000.meta']:
    shutil.copyfile(f'med_max_seq_length512/{suffix}', f'predict/tabfact/model/{suffix}')

max_seq_length = 512
vocab_file = "tapas_medium_model/vocab.txt"
config = tf_example_utils.ClassifierConversionConfig(
    vocab_file=vocab_file,
    max_seq_length=max_seq_length,
    max_column_id=max_seq_length,
    max_row_id=max_seq_length,
    strip_column_names=False,
    add_aggregation_candidates=False,
)
converter = tf_example_utils.ToClassifierTensorflowExample(config)

def convert_interactions_to_examples(tables_and_queries):
    """Calls Tapas converter to convert interaction to example."""
    for idx, (table, queries) in enumerate(tables_and_queries):
        interaction = interaction_pb2.Interaction()
    for position, query in enumerate(queries):
        question = interaction.questions.add()
        question.original_text = query
        question.id = f"{idx}-0_{position}"
    for header in table[0]:
        interaction.table.columns.add().text = header
    for line in table[1:]:
        row = interaction.table.rows.add()
        for cell in line:
            row.cells.add().text = cell
    number_annotation_utils.add_numeric_values(interaction)
    for i in range(len(interaction.questions)):
        try:
            yield converter.convert(interaction, i)
        except ValueError as e:
            print(f"Can't convert interaction: {interaction.id} error: {e}")

def write_tf_example(filename, examples):
    with tf.io.TFRecordWriter(filename) as writer:
        for example in examples:
            writer.write(example.SerializeToString())

def predict(table_data, queries):
    table = [list(map(lambda s: s.strip(), row.split("|"))) 
           for row in table_data.split("\n") if row.strip()]
    examples = convert_interactions_to_examples([(table, queries)])
    write_tf_example("predict/tabfact/tf_examples/test.tfrecord", examples)
    write_tf_example("predict/tabfact/tf_examples/dev.tfrecord", [])

    ! python3 tapas-master/tapas/run_task_main.py \
      --task="TABFACT" \
      --output_dir="predict" \
      --noloop_predict \
      --test_batch_size={len(queries)} \
      --tapas_verbosity="ERROR" \
      --compression_type= \
      --reset_position_index_per_cell \
      --init_checkpoint="tapas_medium_model/model.ckpt" \
      --bert_config_file="tapas_medium_model/bert_config.json" \
      --mode="predict" 2> error

    results_path = "predict/tabfact/model/test.tsv"
    all_results = []
    df = pd.DataFrame(table[1:], columns=table[0])
    display(IPython.display.HTML(df.to_html(index=False)))
    print()
    with open(results_path) as csvfile:
        reader = csv.DictReader(csvfile, delimiter='\t')
    for row in reader:
        supported = int(row["pred_cls"])
        all_results.append(supported)
        score = float(row["logits_cls"])
        position = int(row['position'])
        if supported:
            print("> SUPPORTS:", queries[position])
        else:
            print("> REFUTES:", queries[position])
    return all_results

result = predict("""
tournament            | wins | top - 10 | top - 25 | events | cuts made
masters tournament    | 0    | 0        | 1        | 3      | 2        
us open               | 0    | 0        | 0        | 4      | 3        
the open championship | 0    | 0        | 0        | 2      | 1        
pga championship      | 0    | 1        | 1        | 4      | 2        
totals                | 0    | 1        | 2        | 13     | 8        
""", ["The most frequently occurring number of events is 4", "The most frequently occurring number of events is 3"])

the error would be:

FileNotFoundError                         Traceback (most recent call last)
<ipython-input-44-a676d08da6c0> in <module>
      6 pga championship      | 0    | 1        | 1        | 4      | 2
      7 totals                | 0    | 1        | 2        | 13     | 8
----> 8 """, ["The most frequently occurring number of events is 4", "The most frequently occurring number of events is 3"])

<ipython-input-43-d7d547d8fbe1> in predict(table_data, queries)
     53     display(IPython.display.HTML(df.to_html(index=False)))
     54     print()
---> 55     with open(results_path) as csvfile:
     56         reader = csv.DictReader(csvfile, delimiter='\t')
     57     for row in reader:

FileNotFoundError: [Errno 2] No such file or directory: 'predict/tabfact/model/test.tsv'

This is my directory structure： image image

But when I was running codes on the colab, there are no mistakes. Could you please help me with this issue?

eisenjulian commented 3 years ago

Hi @wcc17864158993 if the tsv is missing it means that the prediction script failed. The logs for it are mapped into a file called error at the root, can you check if you have such a file and share the content?

wcc17864158993 commented 3 years ago

Hello @eisenjulian the file error content:

WARNING:tensorflow:From /opt/conda/lib/python3.6/site-packages/tensorflow/python/compat/v2_compat.py:96: disable_resource_variables (from tensorflow.python.ops.variable_scope) is deprecated and will be removed in a future version.
Instructions for updating:
non-resource variables are not supported in the long term
I1210 14:22:34.871476 140232031508288 run_task_main.py:162] is_built_with_cuda: True
2020-12-10 14:22:34.871764: I tensorflow/core/platform/cpu_feature_guard.cc:143] Your CPU supports instructions that this TensorFlow binary was not compiled to use: SSE4.1 SSE4.2 AVX AVX2 AVX512F FMA
2020-12-10 14:22:34.887337: I tensorflow/core/platform/profile_utils/cpu_utils.cc:102] CPU Frequency: 2300000000 Hz
2020-12-10 14:22:34.894417: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x56221abd8fb0 initialized for platform Host (this does not guarantee that XLA will be used). Devices:
2020-12-10 14:22:34.894504: I tensorflow/compiler/xla/service/service.cc:176]   StreamExecutor device (0): Host, Default Version
2020-12-10 14:22:34.896547: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcuda.so.1
2020-12-10 14:22:34.957768: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1561] Found device 0 with properties: 
pciBusID: 0000:3d:00.0 name: GeForce RTX 2080 Ti computeCapability: 7.5
coreClock: 1.545GHz coreCount: 68 deviceMemorySize: 10.76GiB deviceMemoryBandwidth: 573.69GiB/s
2020-12-10 14:22:34.958055: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.1
2020-12-10 14:22:34.960612: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcublas.so.10
2020-12-10 14:22:34.962806: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcufft.so.10
2020-12-10 14:22:34.963238: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcurand.so.10
2020-12-10 14:22:34.965400: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusolver.so.10
2020-12-10 14:22:34.966510: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusparse.so.10
2020-12-10 14:22:34.971252: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7
2020-12-10 14:22:34.975233: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1703] Adding visible gpu devices: 0
2020-12-10 14:22:34.975331: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.1
2020-12-10 14:22:35.216099: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1102] Device interconnect StreamExecutor with strength 1 edge matrix:
2020-12-10 14:22:35.216160: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1108]      0 
2020-12-10 14:22:35.216171: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1121] 0:   N 
2020-12-10 14:22:35.220232: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1247] Created TensorFlow device (/device:GPU:0 with 10202 MB memory) -> physical GPU (device: 0, name: GeForce RTX 2080 Ti, pci bus id: 0000:3d:00.0, compute capability: 7.5)
2020-12-10 14:22:35.222528: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x56222003b500 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:
2020-12-10 14:22:35.222562: I tensorflow/compiler/xla/service/service.cc:176]   StreamExecutor device (0): GeForce RTX 2080 Ti, Compute Capability 7.5
I1210 14:22:35.223332 140232031508288 run_task_main.py:162] is_gpu_available: True
2020-12-10 14:22:35.225051: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1561] Found device 0 with properties: 
pciBusID: 0000:3d:00.0 name: GeForce RTX 2080 Ti computeCapability: 7.5
coreClock: 1.545GHz coreCount: 68 deviceMemorySize: 10.76GiB deviceMemoryBandwidth: 573.69GiB/s
2020-12-10 14:22:35.225124: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.1
2020-12-10 14:22:35.225152: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcublas.so.10
2020-12-10 14:22:35.225167: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcufft.so.10
2020-12-10 14:22:35.225180: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcurand.so.10
2020-12-10 14:22:35.225194: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusolver.so.10
2020-12-10 14:22:35.225207: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusparse.so.10
2020-12-10 14:22:35.225221: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7
2020-12-10 14:22:35.227685: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1703] Adding visible gpu devices: 0
I1210 14:22:35.227826 140232031508288 run_task_main.py:162] GPUs: [PhysicalDevice(name='/physical_device:GPU:0', device_type='GPU')]
I1210 14:22:35.227957 140232031508288 run_task_main.py:162] Training or predicting ...
I1210 14:22:35.964963 140232031508288 modeling.py:491] position: Tensor("bert/embeddings/ExpandDims:0", shape=(1, 12), dtype=int32)
I1210 14:22:35.973732 140232031508288 modeling.py:493] batched_position: Tensor("bert/embeddings/Repeat/Reshape:0", shape=(?, 12), dtype=int32)
I1210 14:22:35.973974 140232031508288 modeling.py:494] token_type_ids: Tensor("IteratorGetNext:2", shape=(?, 12), dtype=int32)
2020-12-10 14:22:37.752212: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1561] Found device 0 with properties: 
pciBusID: 0000:3d:00.0 name: GeForce RTX 2080 Ti computeCapability: 7.5
coreClock: 1.545GHz coreCount: 68 deviceMemorySize: 10.76GiB deviceMemoryBandwidth: 573.69GiB/s
2020-12-10 14:22:37.752381: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.1
2020-12-10 14:22:37.752419: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcublas.so.10
2020-12-10 14:22:37.752454: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcufft.so.10
2020-12-10 14:22:37.752480: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcurand.so.10
2020-12-10 14:22:37.752501: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusolver.so.10
2020-12-10 14:22:37.752523: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusparse.so.10
2020-12-10 14:22:37.752545: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7
2020-12-10 14:22:37.755123: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1703] Adding visible gpu devices: 0
2020-12-10 14:22:37.755248: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1102] Device interconnect StreamExecutor with strength 1 edge matrix:
2020-12-10 14:22:37.755260: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1108]      0 
2020-12-10 14:22:37.755267: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1121] 0:   N 
2020-12-10 14:22:37.758013: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1247] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 10202 MB memory) -> physical GPU (device: 0, name: GeForce RTX 2080 Ti, pci bus id: 0000:3d:00.0, compute capability: 7.5)
2020-12-10 14:22:39.593654: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcublas.so.10
I1210 14:22:40.108834 140232031508288 modeling.py:491] position: Tensor("bert/embeddings/ExpandDims:0", shape=(1, 12), dtype=int32)
I1210 14:22:40.117084 140232031508288 modeling.py:493] batched_position: Tensor("bert/embeddings/Repeat/Reshape:0", shape=(?, 12), dtype=int32)
I1210 14:22:40.117332 140232031508288 modeling.py:494] token_type_ids: Tensor("IteratorGetNext:2", shape=(?, 12), dtype=int32)
2020-12-10 14:22:41.624194: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1561] Found device 0 with properties: 
pciBusID: 0000:3d:00.0 name: GeForce RTX 2080 Ti computeCapability: 7.5
coreClock: 1.545GHz coreCount: 68 deviceMemorySize: 10.76GiB deviceMemoryBandwidth: 573.69GiB/s
2020-12-10 14:22:41.624314: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.1
2020-12-10 14:22:41.624332: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcublas.so.10
2020-12-10 14:22:41.624347: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcufft.so.10
2020-12-10 14:22:41.624382: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcurand.so.10
2020-12-10 14:22:41.624400: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusolver.so.10
2020-12-10 14:22:41.624417: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusparse.so.10
2020-12-10 14:22:41.624440: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7
2020-12-10 14:22:41.629974: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1703] Adding visible gpu devices: 0
2020-12-10 14:22:41.630030: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1102] Device interconnect StreamExecutor with strength 1 edge matrix:
2020-12-10 14:22:41.630040: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1108]      0 
2020-12-10 14:22:41.630047: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1121] 0:   N 
2020-12-10 14:22:41.637384: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1247] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 10202 MB memory) -> physical GPU (device: 0, name: GeForce RTX 2080 Ti, pci bus id: 0000:3d:00.0, compute capability: 7.5)
2020-12-10 14:22:42.720438: W tensorflow/core/framework/op_kernel.cc:1753] OP_REQUIRES failed at example_parsing_ops.cc:94 : Invalid argument: Key: input_mask.  Can't parse serialized Example.
2020-12-10 14:22:42.720506: W tensorflow/core/framework/op_kernel.cc:1753] OP_REQUIRES failed at example_parsing_ops.cc:94 : Invalid argument: Key: segment_ids.  Can't parse serialized Example.
2020-12-10 14:22:42.720438: W tensorflow/core/framework/op_kernel.cc:1753] OP_REQUIRES failed at example_parsing_ops.cc:94 : Invalid argument: Key: inv_column_ranks.  Can't parse serialized Example.
2020-12-10 14:22:42.720438: W tensorflow/core/framework/op_kernel.cc:1753] OP_REQUIRES failed at example_parsing_ops.cc:94 : Invalid argument: Key: prev_label_ids.  Can't parse serialized Example.
2020-12-10 14:22:42.720438: W tensorflow/core/framework/op_kernel.cc:1753] OP_REQUIRES failed at example_parsing_ops.cc:94 : Invalid argument: Key: numeric_values.  Can't parse serialized Example.
2020-12-10 14:22:42.720464: W tensorflow/core/framework/op_kernel.cc:1753] OP_REQUIRES failed at example_parsing_ops.cc:94 : Invalid argument: Key: row_ids.  Can't parse serialized Example.
2020-12-10 14:22:42.720466: W tensorflow/core/framework/op_kernel.cc:1753] OP_REQUIRES failed at example_parsing_ops.cc:94 : Invalid argument: Key: input_mask.  Can't parse serialized Example.
2020-12-10 14:22:42.720483: W tensorflow/core/framework/op_kernel.cc:1753] OP_REQUIRES failed at example_parsing_ops.cc:94 : Invalid argument: Key: column_ranks.  Can't parse serialized Example.
2020-12-10 14:22:42.720438: W tensorflow/core/framework/op_kernel.cc:1753] OP_REQUIRES failed at example_parsing_ops.cc:94 : Invalid argument: Key: input_mask.  Can't parse serialized Example.
2020-12-10 14:22:42.720522: W tensorflow/core/framework/op_kernel.cc:1753] OP_REQUIRES failed at example_parsing_ops.cc:94 : Invalid argument: Key: segment_ids.  Can't parse serialized Example.
2020-12-10 14:22:42.720572: W tensorflow/core/framework/op_kernel.cc:1753] OP_REQUIRES failed at example_parsing_ops.cc:94 : Invalid argument: Key: prev_label_ids.  Can't parse serialized Example.
2020-12-10 14:22:42.720758: W tensorflow/core/framework/op_kernel.cc:1753] OP_REQUIRES failed at example_parsing_ops.cc:94 : Invalid argument: Key: input_mask.  Can't parse serialized Example.
Traceback (most recent call last):
  File "/opt/conda/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1365, in _do_call
    return fn(*args)
  File "/opt/conda/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1350, in _run_fn
    target_list, run_metadata)
  File "/opt/conda/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1443, in _call_tf_sessionrun
    run_metadata)
tensorflow.python.framework.errors_impl.InvalidArgumentError: 2 root error(s) found.
  (0) Invalid argument: Key: inv_column_ranks.  Can't parse serialized Example.
     [[{{node ParseSingleExample/ParseExample/ParseExampleV2}}]]
     [[IteratorGetNext]]
     [[bert/encoder/Shape/_899]]
  (1) Invalid argument: Key: inv_column_ranks.  Can't parse serialized Example.
     [[{{node ParseSingleExample/ParseExample/ParseExampleV2}}]]
     [[IteratorGetNext]]
0 successful operations.
0 derived errors ignored.

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "tapas-master/tapas/run_task_main.py", line 821, in <module>
    app.run(main)
  File "/opt/conda/lib/python3.6/site-packages/absl/app.py", line 299, in run
    _run_main(main, args)
  File "/opt/conda/lib/python3.6/site-packages/absl/app.py", line 250, in _run_main
    sys.exit(main(argv))
  File "tapas-master/tapas/run_task_main.py", line 806, in main
    loop_predict=FLAGS.loop_predict,
  File "tapas-master/tapas/run_task_main.py", line 504, in _train_and_predict
    global_step=current_step,
  File "tapas-master/tapas/run_task_main.py", line 555, in _predict
    global_step=None,
  File "tapas-master/tapas/run_task_main.py", line 616, in _predict_for_set
    output_token_probabilities=False)
  File "/data-output/tapas-master/tapas/experiments/prediction_utils.py", line 397, in write_predictions
    for prediction in predictions:
  File "/opt/conda/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/tpu/tpu_estimator.py", line 3126, in predict
    rendezvous.raise_errors()
  File "/opt/conda/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/tpu/error_handling.py", line 150, in raise_errors
    six.reraise(typ, value, traceback)
  File "/opt/conda/lib/python3.6/site-packages/six.py", line 703, in reraise
    raise value
  File "/opt/conda/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/tpu/tpu_estimator.py", line 3120, in predict
    yield_single_examples=yield_single_examples):
  File "/opt/conda/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/estimator.py", line 631, in predict
    preds_evaluated = mon_sess.run(predictions)
  File "/opt/conda/lib/python3.6/site-packages/tensorflow/python/training/monitored_session.py", line 778, in run
    run_metadata=run_metadata)
  File "/opt/conda/lib/python3.6/site-packages/tensorflow/python/training/monitored_session.py", line 1283, in run
    run_metadata=run_metadata)
  File "/opt/conda/lib/python3.6/site-packages/tensorflow/python/training/monitored_session.py", line 1384, in run
    raise six.reraise(*original_exc_info)
  File "/opt/conda/lib/python3.6/site-packages/six.py", line 703, in reraise
    raise value
  File "/opt/conda/lib/python3.6/site-packages/tensorflow/python/training/monitored_session.py", line 1369, in run
    return self._sess.run(*args, **kwargs)
  File "/opt/conda/lib/python3.6/site-packages/tensorflow/python/training/monitored_session.py", line 1442, in run
    run_metadata=run_metadata)
  File "/opt/conda/lib/python3.6/site-packages/tensorflow/python/training/monitored_session.py", line 1200, in run
    return self._sess.run(*args, **kwargs)
  File "/opt/conda/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 958, in run
    run_metadata_ptr)
  File "/opt/conda/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1181, in _run
    feed_dict_tensor, options, run_metadata)
  File "/opt/conda/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1359, in _do_run
    run_metadata)
  File "/opt/conda/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1384, in _do_call
    raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.InvalidArgumentError: 2 root error(s) found.
  (0) Invalid argument: Key: inv_column_ranks.  Can't parse serialized Example.
     [[{{node ParseSingleExample/ParseExample/ParseExampleV2}}]]
     [[IteratorGetNext]]
     [[bert/encoder/Shape/_899]]
  (1) Invalid argument: Key: inv_column_ranks.  Can't parse serialized Example.
     [[{{node ParseSingleExample/ParseExample/ParseExampleV2}}]]
     [[IteratorGetNext]]
0 successful operations.
0 derived errors ignored.
eisenjulian commented 3 years ago

Thanks, I'm still not sure the reason for the problem is. It seems that something is wrong in the tf record that was created for prediction, since the inv_column_ranks feature is missing. Can you try doing: 

for example in tf.io.tf_record_iterator("data-output/predict/tabfact/tf_examples/test.tfrecord"):
  print(tf.train.Example.FromString(example))

And the same for dev.tfrecord?

wcc17864158993 commented 3 years ago

@eisenjulian ，I have solved this problem. Thank you for your help！