Giter VIP home page Giter VIP logo

Comments (5)

elmer-garduno avatar elmer-garduno commented on August 23, 2024

To make sure things are fully compatible pleas use TFT version 0.3.0 which we just released yesterday.

https://github.com/tensorflow/transform/releases/tag/v0.3.0

That should help with the InputFnOps problem, please let me know if it works.

from transform.

lukashes avatar lukashes commented on August 23, 2024

Tried this bundle:

tensorflow==1.3
tensorflow_transform==0.3
six==1.10.0

Got the same error:

AttributeError: 'InputFnOps' object has no attribute 'receiver_tensors'

experiment fn:

       estimator = tf.estimator.LinearClassifier(feature_columns=[weighted_reviews],
                                        n_classes=LABEL_SIZE,
                                        model_dir=output_dir,
                                        config=tf.contrib.learn.RunConfig(save_checkpoints_steps=100))

       transformed_metadata = metadata_io.read_metadata(transformed_metadata_dir)
       raw_metadata = metadata_io.read_metadata(raw_metadata_dir)

       train_input_fn = input_fn_maker.build_training_input_fn(
         transformed_metadata,
         transformed_train_filepattern,
         training_batch_size=TRAIN_BATCH_SIZE,
         label_keys=[LABEL_COLUMN])

       eval_input_fn = input_fn_maker.build_training_input_fn(
         transformed_metadata,
         transformed_test_filepattern,
         training_batch_size=1,
         label_keys=[LABEL_COLUMN])

       serving_input_fn = input_fn_maker.build_default_transforming_serving_input_fn(
         raw_metadata=raw_metadata,
         transform_savedmodel_dir=transform_fn_dir + "/transform_fn",
         raw_label_keys=[],
         raw_feature_keys=[KEYWORDS_COLUMN])

       export_strategy = saved_model_export_utils.make_export_strategy(
         serving_input_fn,
         exports_to_keep=5,
         default_output_alternative_key=None)

       max_steps = MAX_STEPS if MAX_STEPS > 0 else \
         TRAIN_NUM_EPOCHS * NUM_TRAIN_INSTANCES / TRAIN_BATCH_SIZE

       return tf.contrib.learn.Experiment(
         estimator=estimator,
         train_steps=max_steps,
         eval_steps=NUM_TEST_INSTANCES,
         train_input_fn=train_input_fn,
         eval_input_fn=eval_input_fn,
         export_strategies=export_strategy,
         min_eval_frequency=5)

from transform.

elmer-garduno avatar elmer-garduno commented on August 23, 2024

I see, yes we need to upload a fix to support the core estimators, will update this bug with more info as soon as we push it to github.

from transform.

davidsoergel avatar davidsoergel commented on August 23, 2024

Yep, that's a known issue; will fix soon.

from transform.

KesterTong avatar KesterTong commented on August 23, 2024

Fixed in 0.3.1

from transform.

Related Issues (20)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.