UnicodeEncodeError: 'charmap' codec can't encode characters in position 1210-1213: character maps to <undefined>

[arbitropy]

--I am trying to run the example evaluation given in this repo on my local machine, but this is the error I am getting. what can I do? below is the whole info.

$python ./question_answering/question_answering.py --model_name_or_path "csebuetnlp/banglabert" --dataset_dir "./question_answering/sample_inputs/" --output_dir "./question_answering/outputs/" --per_device_eval_batch_size=24 --overwrite_output_dir --do_predict D:\Work\anaconda3\envs\banglabert\lib\site-packages\torch\cuda__init__.py:104: UserWarning: NVIDIA GeForce RTX 3050 Laptop GPU with CUDA capability sm_86 is not compatible with the current PyTorch installation. The current PyTorch install supports CUDA capabilities sm_37 sm_50 sm_60 sm_61 sm_70 sm_75 compute_37. If you want to use the NVIDIA GeForce RTX 3050 Laptop GPU GPU with PyTorch, please check the instructions at https://pytorch.org/get-started/locally/

warnings.warn(incompatible_device_warn.format(device_name, capability, " ".join(arch_list), device_name)) 10/02/2023 23:42:15 - WARNING - main - Process rank: -1, device: cuda:0, n_gpu: 1distributed training: False, 16-bits training: False 10/02/2023 23:42:15 - INFO - main - Training/evaluation parameters TrainingArguments( _n_gpu=1, adafactor=False, adam_beta1=0.9, adam_beta2=0.999, adam_epsilon=1e-08, dataloader_drop_last=False, dataloader_num_workers=0, dataloader_pin_memory=True, ddp_find_unused_parameters=None, debug=[], deepspeed=None, disable_tqdm=False, do_eval=False, do_predict=True, do_train=False, eval_accumulation_steps=None, eval_steps=None, evaluation_strategy=IntervalStrategy.NO, fp16=False, fp16_backend=auto, fp16_full_eval=False, fp16_opt_level=O1, gradient_accumulation_steps=1, greater_is_better=None, group_by_length=False, ignore_data_skip=False, label_names=None, label_smoothing_factor=0.0, learning_rate=5e-05, length_column_name=length, load_best_model_at_end=False, local_rank=-1, log_level=-1, log_level_replica=-1, log_on_each_node=True, logging_dir=./question_answering/outputs/runs\Oct02_23-42-14_WIN-849LR3KF8F4, logging_first_step=False, logging_steps=500, logging_strategy=IntervalStrategy.STEPS, lr_scheduler_type=SchedulerType.LINEAR, max_grad_norm=1.0, max_steps=-1, metric_for_best_model=None, mp_parameters=, no_cuda=False, num_train_epochs=3.0, output_dir=./question_answering/outputs/, overwrite_output_dir=True, past_index=-1, per_device_eval_batch_size=24, per_device_train_batch_size=8, prediction_loss_only=False, push_to_hub=False, push_to_hub_model_id=outputs, push_to_hub_organization=None, push_to_hub_token=None, remove_unused_columns=True, report_to=[], resume_from_checkpoint=None, run_name=./question_answering/outputs/, save_on_each_node=False, save_steps=500, save_strategy=IntervalStrategy.STEPS, save_total_limit=None, seed=42, sharded_ddp=[], skip_memory_metrics=True, tpu_metrics_debug=False, tpu_num_cores=None, use_legacy_prediction_loop=False, warmup_ratio=0.0, warmup_steps=0, weight_decay=0.0, ) 10/02/2023 23:42:15 - WARNING - datasets.builder - Using custom data configuration default-431d55e0f1c961a4 10/02/2023 23:42:15 - INFO - datasets.builder - Overwrite dataset info from restored data version. 10/02/2023 23:42:15 - INFO - datasets.info - Loading Dataset info from C:\Users\Administrator.cache\huggingface\datasets\qa_dataset_builder\default-431d55e0f1c961a4\0.0.0 10/02/2023 23:42:15 - WARNING - datasets.builder - Reusing dataset qa_dataset_builder (C:\Users\Administrator.cache\huggingface\datasets\qa_dataset_builder\default-431d55e0f1c961a4\0.0.0) 10/02/2023 23:42:15 - INFO - datasets.info - Loading Dataset info from C:\Users\Administrator.cache\huggingface\datasets\qa_dataset_builder\default-431d55e0f1c961a4\0.0.0 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 3/3 [00:00<00:00, 57.23it/s] [INFO|configuration_utils.py:561] 2023-10-02 23:42:16,323 >> loading configuration file https://huggingface.co/csebuetnlp/banglabert/resolve/main/config.json from cache at C:\Users\Administrator/.cache\huggingface\transformers\60928dc4b87f5881692890e6541e6538f91588d2ea40cbbbdc04cfb2cb83a6b1.2388211ba94f448fcf40aef3c9526142a8c2f2a8fb4fce8a3801462f51b2bab5 [INFO|configuration_utils.py:598] 2023-10-02 23:42:16,326 >> Model config ElectraConfig { "architectures": [ "ElectraForPreTraining" ], "attention_probs_dropout_prob": 0.1, "classifier_dropout": null, "embedding_size": 768, "hidden_act": "gelu", "hidden_dropout_prob": 0.1, "hidden_size": 768, "initializer_range": 0.02, "intermediate_size": 3072, "layer_norm_eps": 1e-12, "max_position_embeddings": 512, "model_type": "electra", "num_attention_heads": 12, "num_hidden_layers": 12, "pad_token_id": 0, "position_embedding_type": "absolute", "summary_activation": "gelu", "summary_last_dropout": 0.1, "summary_type": "first", "summary_use_proj": true, "transformers_version": "4.11.0.dev0", "type_vocab_size": 2, "vocab_size": 32000 }

[INFO|configuration_utils.py:561] 2023-10-02 23:42:17,541 >> loading configuration file https://huggingface.co/csebuetnlp/banglabert/resolve/main/config.json from cache at C:\Users\Administrator/.cache\huggingface\transformers\60928dc4b87f5881692890e6541e6538f91588d2ea40cbbbdc04cfb2cb83a6b1.2388211ba94f448fcf40aef3c9526142a8c2f2a8fb4fce8a3801462f51b2bab5 [INFO|configuration_utils.py:598] 2023-10-02 23:42:17,544 >> Model config ElectraConfig { "architectures": [ "ElectraForPreTraining" ], "attention_probs_dropout_prob": 0.1, "classifier_dropout": null, "embedding_size": 768, "hidden_act": "gelu", "hidden_dropout_prob": 0.1, "hidden_size": 768, "initializer_range": 0.02, "intermediate_size": 3072, "layer_norm_eps": 1e-12, "max_position_embeddings": 512, "model_type": "electra", "num_attention_heads": 12, "num_hidden_layers": 12, "pad_token_id": 0, "position_embedding_type": "absolute", "summary_activation": "gelu", "summary_last_dropout": 0.1, "summary_type": "first", "summary_use_proj": true, "transformers_version": "4.11.0.dev0", "type_vocab_size": 2, "vocab_size": 32000 }

[INFO|tokenization_utils_base.py:1739] 2023-10-02 23:42:23,760 >> loading file https://huggingface.co/csebuetnlp/banglabert/resolve/main/vocab.txt from cache at C:\Users\Administrator/.cache\huggingface\transformers\65e95b847336b6bf69b37fdb8682a97e822799adcd9745dcf9bf44cfe4db1b9a.8f92ca2cf7e2eaa550b10c40331ae9bf0f2e40abe3b549f66a3d7f13bfc6de47 [INFO|tokenization_utils_base.py:1739] 2023-10-02 23:42:23,760 >> loading file https://huggingface.co/csebuetnlp/banglabert/resolve/main/tokenizer.json from cache at None [INFO|tokenization_utils_base.py:1739] 2023-10-02 23:42:23,760 >> loading file https://huggingface.co/csebuetnlp/banglabert/resolve/main/added_tokens.json from cache at None [INFO|tokenization_utils_base.py:1739] 2023-10-02 23:42:23,761 >> loading file https://huggingface.co/csebuetnlp/banglabert/resolve/main/special_tokens_map.json from cache at C:\Users\Administrator/.cache\huggingface\transformers\7820dfc553e8dfb8a1e82042b7d0d691c7a7cd1e30ed2974218f696e81c5f3b1.dd8bd9bfd3664b530ea4e645105f557769387b3da9f79bdb55ed556bdd80611d [INFO|tokenization_utils_base.py:1739] 2023-10-02 23:42:23,761 >> loading file https://huggingface.co/csebuetnlp/banglabert/resolve/main/tokenizer_config.json from cache at C:\Users\Administrator/.cache\huggingface\transformers\76fa87a0ec9c34c9b15732bf7e06bced447feff46287b8e7d246a55d301784d7.b4f59cefeba4296760d2cf1037142788b96f2be40230bf6393d2fba714562485 [INFO|configuration_utils.py:561] 2023-10-02 23:42:25,458 >> loading configuration file https://huggingface.co/csebuetnlp/banglabert/resolve/main/config.json from cache at C:\Users\Administrator/.cache\huggingface\transformers\60928dc4b87f5881692890e6541e6538f91588d2ea40cbbbdc04cfb2cb83a6b1.2388211ba94f448fcf40aef3c9526142a8c2f2a8fb4fce8a3801462f51b2bab5 [INFO|configuration_utils.py:598] 2023-10-02 23:42:25,460 >> Model config ElectraConfig { "architectures": [ "ElectraForPreTraining" ], "attention_probs_dropout_prob": 0.1, "classifier_dropout": null, "embedding_size": 768, "hidden_act": "gelu", "hidden_dropout_prob": 0.1, "hidden_size": 768, "initializer_range": 0.02, "intermediate_size": 3072, "layer_norm_eps": 1e-12, "max_position_embeddings": 512, "model_type": "electra", "num_attention_heads": 12, "num_hidden_layers": 12, "pad_token_id": 0, "position_embedding_type": "absolute", "summary_activation": "gelu", "summary_last_dropout": 0.1, "summary_type": "first", "summary_use_proj": true, "transformers_version": "4.11.0.dev0", "type_vocab_size": 2, "vocab_size": 32000 }

[INFO|configuration_utils.py:561] 2023-10-02 23:42:26,158 >> loading configuration file https://huggingface.co/csebuetnlp/banglabert/resolve/main/config.json from cache at C:\Users\Administrator/.cache\huggingface\transformers\60928dc4b87f5881692890e6541e6538f91588d2ea40cbbbdc04cfb2cb83a6b1.2388211ba94f448fcf40aef3c9526142a8c2f2a8fb4fce8a3801462f51b2bab5 [INFO|configuration_utils.py:598] 2023-10-02 23:42:26,161 >> Model config ElectraConfig { "architectures": [ "ElectraForPreTraining" ], "attention_probs_dropout_prob": 0.1, "classifier_dropout": null, "embedding_size": 768, "hidden_act": "gelu", "hidden_dropout_prob": 0.1, "hidden_size": 768, "initializer_range": 0.02, "intermediate_size": 3072, "layer_norm_eps": 1e-12, "max_position_embeddings": 512, "model_type": "electra", "num_attention_heads": 12, "num_hidden_layers": 12, "pad_token_id": 0, "position_embedding_type": "absolute", "summary_activation": "gelu", "summary_last_dropout": 0.1, "summary_type": "first", "summary_use_proj": true, "transformers_version": "4.11.0.dev0", "type_vocab_size": 2, "vocab_size": 32000 }

[INFO|file_utils.py:1665] 2023-10-02 23:42:26,817 >> https://huggingface.co/csebuetnlp/banglabert/resolve/main/pytorch_model.bin not found in cache or force_download set to True, downloading to C:\Users\Administrator.cache\huggingface\transformers\tmpe23b9o4n Downloading: 100%|██████████████████████████████████████████████████████████████████████████████████████████████| 443M/443M [05:26<00:00, 1.36MB/s] [INFO|file_utils.py:1669] 2023-10-02 23:47:54,041 >> storing https://huggingface.co/csebuetnlp/banglabert/resolve/main/pytorch_model.bin in cache at C:\Users\Administrator/.cache\huggingface\transformers\913ea71768a80ccdde3a9ab9a88cf2a93f37a52008896997655d0f63b0d0743a.8aaedac281b72dbb5296319c53be5a4e4a52339eded3f68d49201e140e221615 [INFO|file_utils.py:1677] 2023-10-02 23:47:54,042 >> creating metadata file for C:\Users\Administrator/.cache\huggingface\transformers\913ea71768a80ccdde3a9ab9a88cf2a93f37a52008896997655d0f63b0d0743a.8aaedac281b72dbb5296319c53be5a4e4a52339eded3f68d49201e140e221615 [INFO|modeling_utils.py:1279] 2023-10-02 23:47:54,044 >> loading weights file https://huggingface.co/csebuetnlp/banglabert/resolve/main/pytorch_model.bin from cache at C:\Users\Administrator/.cache\huggingface\transformers\913ea71768a80ccdde3a9ab9a88cf2a93f37a52008896997655d0f63b0d0743a.8aaedac281b72dbb5296319c53be5a4e4a52339eded3f68d49201e140e221615 [WARNING|modeling_utils.py:1516] 2023-10-02 23:47:54,822 >> Some weights of the model checkpoint at csebuetnlp/banglabert were not used when initializing ElectraForQuestionAnswering: ['discriminator_predictions.dense_prediction.weight', 'discriminator_predictions.dense.weight', 'discriminator_predictions.dense.bias', 'discriminator_predictions.dense_prediction.bias']

• This IS expected if you are initializing ElectraForQuestionAnswering from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).
• This IS NOT expected if you are initializing ElectraForQuestionAnswering from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). [WARNING|modeling_utils.py:1527] 2023-10-02 23:47:54,823 >> Some weights of ElectraForQuestionAnswering were not initialized from the model checkpoint at csebuetnlp/banglabert and are newly initialized: ['qa_outputs.bias', 'qa_outputs.weight'] You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference. 10/02/2023 23:47:54 - WARNING - datasets.fingerprint - Parameter 'function'=<function main..normalize_example at 0x00000239F12A9EE8> of the transform datasets.arrow_dataset.Dataset._map_single couldn't be hashed properly, a random hash was used instead. Make sure your transforms and parameters are serializable with pickle or dill for the dataset fingerprinting and caching to work. If you reuse this transform, the caching mechanism will consider it to be different from the previous calls and recompute everything. This warning is only showed once. Subsequent hashing failures won't be showed. 10/02/2023 23:47:54 - WARNING - datasets.arrow_dataset - Loading cached processed dataset at C:\Users\Administrator.cache\huggingface\datasets\qa_dataset_builder\default-431d55e0f1c961a4\0.0.0\cache-1c80317fa3b1799d.arrow 10/02/2023 23:47:54 - INFO - datasets.fingerprint - Parameter 'function'=<function main..normalize_example at 0x00000239F12A9EE8> of the transform datasets.arrow_dataset.Dataset._map_single couldn't be hashed properly, a random hash was used instead. 10/02/2023 23:47:54 - WARNING - datasets.arrow_dataset - Loading cached processed dataset at C:\Users\Administrator.cache\huggingface\datasets\qa_dataset_builder\default-431d55e0f1c961a4\0.0.0\cache-bdd640fb06671ad1.arrow 10/02/2023 23:47:54 - INFO - datasets.fingerprint - Parameter 'function'=<function main..normalize_example at 0x00000239F12A9EE8> of the transform datasets.arrow_dataset.Dataset._map_single couldn't be hashed properly, a random hash was used instead. 10/02/2023 23:47:54 - WARNING - datasets.arrow_dataset - Loading cached processed dataset at C:\Users\Administrator.cache\huggingface\datasets\qa_dataset_builder\default-431d55e0f1c961a4\0.0.0\cache-3eb13b9046685257.arrow 10/02/2023 23:47:54 - INFO - datasets.fingerprint - Parameter 'function'=<function main..prepare_validation_features at 0x00000239F11B7828> of the transform datasets.arrow_dataset.Dataset._map_single couldn't be hashed properly, a random hash was used instead. 10/02/2023 23:47:54 - WARNING - datasets.arrow_dataset - Loading cached processed dataset at C:\Users\Administrator.cache\huggingface\datasets\qa_dataset_builder\default-431d55e0f1c961a4\0.0.0\cache-23b8c1e9392456de.arrow 10/02/2023 23:47:58 - INFO - datasets.load - Found main folder for metric https://raw.githubusercontent.com/huggingface/datasets/1.11.0/metrics/squad/squad.py at C:\Users\Administrator.cache\huggingface\modules\datasets_modules\metrics\squad 10/02/2023 23:47:58 - INFO - datasets.load - Found specific version folder for metric https://raw.githubusercontent.com/huggingface/datasets/1.11.0/metrics/squad/squad.py at C:\Users\Administrator.cache\huggingface\modules\datasets_modules\metrics\squad\513bf9facd7f12b0871a3d74c6999c866ce28196c9cdb151dcf934848655d77e 10/02/2023 23:47:58 - INFO - datasets.load - Found script file from https://raw.githubusercontent.com/huggingface/datasets/1.11.0/metrics/squad/squad.py to C:\Users\Administrator.cache\huggingface\modules\datasets_modules\metrics\squad\513bf9facd7f12b0871a3d74c6999c866ce28196c9cdb151dcf934848655d77e\squad.py 10/02/2023 23:47:58 - INFO - datasets.load - Couldn't find dataset infos file at https://raw.githubusercontent.com/huggingface/datasets/1.11.0/metrics/squad/dataset_infos.json 10/02/2023 23:47:58 - INFO - datasets.load - Found metadata file for metric https://raw.githubusercontent.com/huggingface/datasets/1.11.0/metrics/squad/squad.py at C:\Users\Administrator.cache\huggingface\modules\datasets_modules\metrics\squad\513bf9facd7f12b0871a3d74c6999c866ce28196c9cdb151dcf934848655d77e\squad.json 10/02/2023 23:47:58 - INFO - datasets.load - Found local import file from C:\Users\Administrator.cache\huggingface\datasets\downloads\188f3c7a325773b47d41f3e0f7ab9fd3cb20e597010b3a9d780c878dacc10ce3.6f69c3ff9e10aa1cbdc6e91d27e158ea86a785f54a36a9e964ef8b3b78cf3cd6.py at C:\Users\Administrator.cache\huggingface\modules\datasets_modules\metrics\squad\513bf9facd7f12b0871a3d74c6999c866ce28196c9cdb151dcf934848655d77e\evaluate.py 10/03/2023 00:11:50 - INFO - main - Predict [INFO|trainer.py:521] 2023-10-03 00:11:50,847 >> The following columns in the test set don't have a corresponding argument in ElectraForQuestionAnswering.forward and have been ignored: offset_mapping, example_id. [INFO|trainer.py:2181] 2023-10-03 00:11:50,932 >> Running Prediction [INFO|trainer.py:2183] 2023-10-03 00:11:50,932 >> Num examples = 2614 [INFO|trainer.py:2186] 2023-10-03 00:11:50,933 >> Batch size = 24 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████| 109/109 [05:48<00:00, 3.16s/it]10/03/2023 00:19:02 - INFO - utils - Post-processing 2504 example predictions split into 2614 features. 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████| 2504/2504 [00:04<00:00, 504.38it/s] 10/03/2023 00:19:07 - INFO - utils - Saving predictions to ./question_answering/outputs/predict_predictions.json.493/2504 [00:04<00:00, 541.44it/s] Traceback (most recent call last): File "./question_answering/question_answering.py", line 617, in main() File "./question_answering/question_answering.py", line 599, in main results = trainer.predict(predict_dataset, predict_examples) File "D:\Work\anaconda3\envs\banglabert\question_answering\utils.py", line 426, in predict predictions = self.post_process_function(predict_examples, predict_dataset, output.predictions, "predict") File "./question_answering/question_answering.py", line 534, in post_processing_function prefix=stage, File "D:\Work\anaconda3\envs\banglabert\question_answering\utils.py", line 345, in postprocess_qa_predictions writer.write(json.dumps(all_predictions, ensure_ascii=False, indent=4) + "\n") File "D:\Work\anaconda3\envs\banglabert\lib\encodings\cp1252.py", line 19, in encode return codecs.charmap_encode(input,self.errors,encoding_table)[0] UnicodeEncodeError: 'charmap' codec can't encode characters in position 1210-1213: character maps to 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████| 109/109 [05:55<00:00, 3.26s/it]

[abhik1505040]

Your Python installation seems to be using windows-1252 encoding by default instead of utf-8. Try invoking your Python scripts like python -X utf8 <path_to_script> <arguments> to see whether it fixes the issue.

[arbitropy]

This fixed it, but now I get a new error. 10/03/2023 22:21:08 - INFO - __main__ - *** Predict *** [INFO|trainer.py:521] 2023-10-03 22:21:08,957 >> The following columns in the test set don't have a corresponding argument inElectraForQuestionAnswering.forwardand have been ignored: example_id, offset_mapping. [INFO|trainer.py:2181] 2023-10-03 22:21:09,046 >> ***** Running Prediction ***** [INFO|trainer.py:2183] 2023-10-03 22:21:09,047 >> Num examples = 2614 [INFO|trainer.py:2186] 2023-10-03 22:21:09,047 >> Batch size = 24 100%|██| 109/109 [05:48<00:00, 3.17s/it]10/03/2023 22:28:21 - INFO - utils - Post-processing 2504 example predictions split into 2614 features. 100%|█| 2504/2504 [00:05<00:00, 492.78it/ 10/03/2023 22:28:26 - INFO - utils - Saving predictions to ./question_answering/outputs/predict_predictions.json. 10/03/2023 22:28:26 - INFO - utils - Saving nbest_preds to ./question_answering/outputs/predict_nbest_predictions.json. Traceback (most recent call last): File "./question_answering/question_answering.py", line 617, in <module> main() File "./question_answering/question_answering.py", line 599, in main results = trainer.predict(predict_dataset, predict_examples) File "D:\Work\anaconda3\envs\banglabert\question_answering\utils.py", line 427, in predict metrics = self.compute_metrics(predictions) File "./question_answering/question_answering.py", line 550, in compute_metrics return metric.compute(predictions=p.predictions, references=p.label_ids) 196c9cdb151dcf934848655d77e\evaluate.py", line 67, in evaluate exact_match += metric_max_over_ground_truths(exact_match_score, prediction, ground_truths) File "C:\Users\Administrator\.cache\huggingface\modules\datasets_modules\metrics\squad\513bf9facd7f12b0871a3d74c6999c866ce28196c9cdb151dcf934848655d77e\evaluate.py", line 52, in metric_max_over_ground_truths return max(scores_for_ground_truths) ValueError: max() arg is an empty sequence100%|██| 109/109 [05:56<00:00, 3.27s/it]

Edit: The output JSON files were correctly generated, in the folder. I can't seem to understand what's not being added.

[abhik1505040]

If you are using the sample dataset provided with this repo, you need to add the flag --allow_null_ans