AssertionError: Some Python objects were not bound to checkpointed values, likely due to changes in the Python program #9254
Description
Prerequisites
Please answer the following questions for yourself before submitting an issue.
- [X ] I am using the latest TensorFlow Model Garden release and TensorFlow 2.
- [ X] I am reporting the issue to the correct repository. (Model Garden official or research directory)
- [X ] I checked to make sure that this issue has not been filed already.
1. The entire URL of the file you are using
https://github.com/tensorflow/models/official/nlp/albert/run_classifier.py
2. Describe the bug
After finetuning ALBERT on the MNLI task I change the mode to 'predict' in the run_classifier.py
and receive the error:
AssertionError: Some Python objects were not bound to checkpointed values, likely due to changes in the Python program
3. Steps to reproduce
Running predict after fine-tuning on MNLI and the finetuned models are present in finetune_results
export MODEL_DIR=finetune_results
export GLUE_DIR=output_glue_data
export TASK=MNLI
python run_classifier.py
--mode='predict'
--input_meta_data_path=${GLUE_DIR}/${TASK}_meta_data
--train_data_path=${GLUE_DIR}/${TASK}_train.tf_record
--eval_data_path=${GLUE_DIR}/${TASK}_eval.tf_record
--bert_config_file=albert_xxlarge/albert_config.json
--train_batch_size=4
--eval_batch_size=4
--steps_per_loop=1
--learning_rate=2e-5
--num_train_epochs=3
--model_dir=${MODEL_DIR}
--distribution_strategy=one_device
4. Expected behavior
An output file labeled test_results.tsv
5. Additional context
6. System information
- OS Platform and Distribution (e.g., Linux Ubuntu 16.04): Linux 18.04
- TensorFlow installed from (source or binary): binary
- TensorFlow version (use command below): 2.4.0-dev20200908
- Python version:3.7.9
- CUDA/cuDNN version:11.0/8.0.1
- GPU model and memory: Tesla V100 16 GB