How can i run a pretrained model for prediction on new data?

[aravindchaluvadi]

Can you guys suggest me what pipeline should i follow to run a pretrained model of the squad version using the tar files on a new text data?

python -m albert.run_squad_v1 \
  --albert_config_file=... \
  --output_dir=... \
  --train_file=... \
  --predict_file=... \
  --train_feature_file=... \
  --predict_feature_file=... \
  --predict_feature_left_file=... \
  --init_checkpoint=... \
  --spm_model_file=... \
  --do_lower_case \
  --max_seq_length=384 \
  --doc_stride=128 \
  --max_query_length=64 \
  --do_train=true \
  --do_predict=true \
  --train_batch_size=48 \
  --predict_batch_size=8 \
  --learning_rate=5e-5 \
  --num_train_epochs=2.0 \
  --warmup_proportion=.1 \
  --save_checkpoints_steps=5000 \
  --n_best_size=20 \
  --max_answer_length=30

[aravindchaluvadi]

The Below is the command that Iam trying to execute:

python -m run_squad_v1 \
  --albert_config_file=/media/xxxx/NewVolume/ALBERT/albert_base/albert_config.json \
  --output_dir=/media/xxxx/NewVolume/ALBERT/tmp \
  --train_file=/media/xxxx/NewVolume/ALBERT/data1/train-v1.1.json \
  --predict_file=/media/xxxx/NewVolume/ALBERT/data1/dev-v1.1.json \
  --train_feature_file=... \
  --predict_feature_file=... \
  --predict_feature_left_file=... \
  --init_checkpoint=... \
  --spm_model_file=/media/xxxx/NewVolume/ALBERT/albert_base/30k-clean.model \
  --do_lower_case \
  --max_seq_length=384 \
  --doc_stride=128 \
  --max_query_length=64 \
  --do_train=false \
  --do_predict=true \
  --train_batch_size=48 \
  --predict_batch_size=8 \
  --learning_rate=5e-5 \
  --num_train_epochs=2.0 \
  --warmup_proportion=.1 \
  --save_checkpoints_steps=5000 \
  --n_best_size=20 \
  --max_answer_length=30

[aravindchaluvadi]

Below is the error I am facing while running the command:

 python -m run_squad_v1 \
>   --albert_config_file=/media/xxxx/NewVolume/ALBERT/albert_base/albert_config.json \
>   --output_dir=/media/xxxx/NewVolume/ALBERT/tmp \
>   --train_file=/media/xxxx/NewVolume/ALBERT/data1/train-v1.1.json \
>   --predict_file=/media/xxxx/NewVolume/ALBERT/data1/dev-v1.1.json \
>   --spm_model_file=/media/xxxx/NewVolume/ALBERT/albert_base/30k-clean.model \
>   --do_lower_case \
>   --max_seq_length=384 \
>   --doc_stride=128 \
>   --max_query_length=64 \
>   --do_train=false \
>   --do_predict=true \
>   --train_batch_size=48 \
>   --predict_batch_size=8 \
>   --learning_rate=5e-5 \
>   --num_train_epochs=2.0 \
>   --warmup_proportion=.1 \
>   --save_checkpoints_steps=5000 \
>   --n_best_size=20 \
>   --max_answer_length=30
WARNING:tensorflow:From /media/xxxx/NewVolume/ALBERT/run_squad_v1.py:206: The name tf.gfile.MakeDirs is deprecated. Please use tf.io.gfile.makedirs instead.

W0113 15:12:16.637617 140307062036288 module_wrapper.py:139] From /media/xxxx/NewVolume/ALBERT/run_squad_v1.py:206: The name tf.gfile.MakeDirs is deprecated. Please use tf.io.gfile.makedirs instead.

INFO:tensorflow:loading sentence piece model
I0113 15:12:16.637814 140307062036288 tokenization.py:240] loading sentence piece model
WARNING:tensorflow:Estimator's model_fn (<function v1_model_fn_builder.<locals>.model_fn at 0x7f9b633440d0>) includes params argument, but params are not passed to Estimator.
W0113 15:12:17.200998 140307062036288 estimator.py:1994] Estimator's model_fn (<function v1_model_fn_builder.<locals>.model_fn at 0x7f9b633440d0>) includes params argument, but params are not passed to Estimator.
INFO:tensorflow:Using config: {'_model_dir': '/media/xxxx/NewVolume/ALBERT/tmp', '_tf_random_seed': None, '_save_summary_steps': 100, '_save_checkpoints_steps': 5000, '_save_checkpoints_secs': None, '_session_config': allow_soft_placement: true
graph_options {
  rewrite_options {
    meta_optimizer_iterations: ONE
  }
}
, '_keep_checkpoint_max': 0, '_keep_checkpoint_every_n_hours': 10000, '_log_step_count_steps': None, '_train_distribute': None, '_device_fn': None, '_protocol': None, '_eval_distribute': None, '_experimental_distribute': None, '_experimental_max_worker_delay_secs': None, '_session_creation_timeout_secs': 7200, '_service': None, '_cluster_spec': <tensorflow.python.training.server_lib.ClusterSpec object at 0x7f9b66fd70b8>, '_task_type': 'worker', '_task_id': 0, '_global_id_in_cluster': 0, '_master': '', '_evaluation_master': '', '_is_chief': True, '_num_ps_replicas': 0, '_num_worker_replicas': 1, '_tpu_config': TPUConfig(iterations_per_loop=1000, num_shards=8, num_cores_per_replica=None, per_host_input_for_training=3, tpu_job_name=None, initial_infeed_sleep_secs=None, input_partition_dims=None, eval_training_input_configuration=2, experimental_host_call_every_n_steps=1), '_cluster': None}
I0113 15:12:17.201757 140307062036288 estimator.py:212] Using config: {'_model_dir': '/media/xxxx/NewVolume/ALBERT/tmp', '_tf_random_seed': None, '_save_summary_steps': 100, '_save_checkpoints_steps': 5000, '_save_checkpoints_secs': None, '_session_config': allow_soft_placement: true
graph_options {
  rewrite_options {
    meta_optimizer_iterations: ONE
  }
}
, '_keep_checkpoint_max': 0, '_keep_checkpoint_every_n_hours': 10000, '_log_step_count_steps': None, '_train_distribute': None, '_device_fn': None, '_protocol': None, '_eval_distribute': None, '_experimental_distribute': None, '_experimental_max_worker_delay_secs': None, '_session_creation_timeout_secs': 7200, '_service': None, '_cluster_spec': <tensorflow.python.training.server_lib.ClusterSpec object at 0x7f9b66fd70b8>, '_task_type': 'worker', '_task_id': 0, '_global_id_in_cluster': 0, '_master': '', '_evaluation_master': '', '_is_chief': True, '_num_ps_replicas': 0, '_num_worker_replicas': 1, '_tpu_config': TPUConfig(iterations_per_loop=1000, num_shards=8, num_cores_per_replica=None, per_host_input_for_training=3, tpu_job_name=None, initial_infeed_sleep_secs=None, input_partition_dims=None, eval_training_input_configuration=2, experimental_host_call_every_n_steps=1), '_cluster': None}
INFO:tensorflow:_TPUContext: eval_on_tpu True
I0113 15:12:17.202082 140307062036288 tpu_context.py:220] _TPUContext: eval_on_tpu True
WARNING:tensorflow:eval_on_tpu ignored because use_tpu is False.
W0113 15:12:17.202302 140307062036288 tpu_context.py:222] eval_on_tpu ignored because use_tpu is False.
WARNING:tensorflow:From /media/xxxx/NewVolume/ALBERT/run_squad_v1.py:303: The name tf.gfile.Open is deprecated. Please use tf.io.gfile.GFile instead.

W0113 15:12:17.202427 140307062036288 module_wrapper.py:139] From /media/xxxx/NewVolume/ALBERT/run_squad_v1.py:303: The name tf.gfile.Open is deprecated. Please use tf.io.gfile.GFile instead.

WARNING:tensorflow:From /media/xxxx/NewVolume/ALBERT/run_squad_v1.py:309: The name tf.gfile.Exists is deprecated. Please use tf.io.gfile.exists instead.

W0113 15:12:17.317886 140307062036288 module_wrapper.py:139] From /media/xxxx/NewVolume/ALBERT/run_squad_v1.py:309: The name tf.gfile.Exists is deprecated. Please use tf.io.gfile.exists instead.

Traceback (most recent call last):
  File "/home/xxxx/anaconda3/envs/albert/lib/python3.6/runpy.py", line 193, in _run_module_as_main
    "__main__", mod_spec)
  File "/home/xxxx/anaconda3/envs/albert/lib/python3.6/runpy.py", line 85, in _run_code
    exec(code, run_globals)
  File "/media/xxxx/NewVolume/ALBERT/run_squad_v1.py", line 478, in <module>
    tf.compat.v1.app.run()
  File "/home/xxxx/.local/lib/python3.6/site-packages/tensorflow_core/python/platform/app.py", line 40, in run
    _run(main=main, argv=argv, flags_parser=_parse_flags_tolerate_undef)
  File "/home/xxxx/.local/lib/python3.6/site-packages/absl/app.py", line 299, in run
    _run_main(main, args)
  File "/home/xxxx/.local/lib/python3.6/site-packages/absl/app.py", line 250, in _run_main
    sys.exit(main(argv))
  File "/media/xxxx/NewVolume/ALBERT/run_squad_v1.py", line 309, in main
    if (tf.gfile.Exists(FLAGS.predict_feature_file) and tf.gfile.Exists(
  File "/home/xxxx/.local/lib/python3.6/site-packages/tensorflow_core/python/lib/io/file_io.py", line 262, in file_exists
    return file_exists_v2(filename)
  File "/home/xxxx/.local/lib/python3.6/site-packages/tensorflow_core/python/lib/io/file_io.py", line 280, in file_exists_v2
    pywrap_tensorflow.FileExists(compat.as_bytes(path))
  File "/home/xxxx/.local/lib/python3.6/site-packages/tensorflow_core/python/util/compat.py", line 71, in as_bytes
    (bytes_or_text,))
TypeError: Expected binary or unicode string, got None

@google-research

[Bidek56]

This is simple, you need to add: --predict_feature_file="$SQUAD_DIR/dev.tfrecord"

[hejunqing]

where did you get the 30k-clean.model ?