Upload 9 files

README.md

@@ -0,0 +1,10 @@
+## TextAttack Model Card
+This `bert-base-uncased` model was fine-tuned for sequence classification using TextAttack 
+and the yelp_polarity dataset loaded using the `nlp` library. The model was fine-tuned 
+for 5 epochs with a batch size of 16, a learning 
+rate of 5e-05, and a maximum sequence length of 256. 
+Since this was a classification task, the model was trained with a cross-entropy loss function. 
+The best score the model achieved on this task was 0.9699473684210527, as measured by the 
+eval set accuracy, found after 4 epochs.
+
+For more information, check out [TextAttack on Github](https://github.com/QData/TextAttack).

config.json

@@ -0,0 +1,21 @@
+{
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "finetuning_task": "yelp_polarity",
+  "gradient_checkpointing": false,
+  "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": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "type_vocab_size": 2,
+  "vocab_size": 30522
+}

flax_model.msgpack

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80c93148809240ce872694c420e0b28a6d5048518edd50f91b9ac4f2825be5d5
+size 437942328

log.txt

@@ -0,0 +1,34 @@
+Writing logs to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/bert-base-uncased-yelp_polarity-2020-07-08-10:42/log.txt.
+Loading nlp dataset yelp_polarity, split train.
+Loading nlp dataset yelp_polarity, split test.
+Loaded dataset. Found: 2 labels: ([0, 1])
+Loading transformers AutoModelForSequenceClassification: bert-base-uncased
+Tokenizing training data. (len: 560000)
+Tokenizing eval data (len: 38000)
+Loaded data and tokenized in 720.6436557769775s
+Using torch.nn.DataParallel.
+Training model across 4 GPUs
+Wrote original training args to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/bert-base-uncased-yelp_polarity-2020-07-08-10:42/train_args.json.
+***** Running training *****
+	Num examples = 560000
+	Batch size = 16
+	Max sequence length = 256
+	Num steps = 175000
+	Num epochs = 5
+	Learning rate = 5e-05
+Eval accuracy: 95.95263157894736%
+Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/bert-base-uncased-yelp_polarity-2020-07-08-10:42/.
+Eval accuracy: 96.59473684210526%
+Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/bert-base-uncased-yelp_polarity-2020-07-08-10:42/.
+Eval accuracy: 96.69473684210527%
+Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/bert-base-uncased-yelp_polarity-2020-07-08-10:42/.
+Eval accuracy: 96.91052631578947%
+Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/bert-base-uncased-yelp_polarity-2020-07-08-10:42/.
+Eval accuracy: 96.99473684210527%
+Best acc found. Saved model to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/bert-base-uncased-yelp_polarity-2020-07-08-10:42/.
+Finished training. Re-loading and evaluating model from disk.
+Loading transformers AutoModelForSequenceClassification: bert-base-uncased
+Eval of saved model accuracy: 96.99473684210527%
+Saved tokenizer <textattack.models.tokenizers.auto_tokenizer.AutoTokenizer object at 0x7fcc548eb730> to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/bert-base-uncased-yelp_polarity-2020-07-08-10:42/.
+Wrote README to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/bert-base-uncased-yelp_polarity-2020-07-08-10:42/README.md.
+Wrote final training args to /p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/bert-base-uncased-yelp_polarity-2020-07-08-10:42/train_args.json.

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:934989a6c6465c39311acbb466e8e27e4a715539c3baa95c7c66b424a6c27a66
+size 437985387

special_tokens_map.json

@@ -0,0 +1 @@
+{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}

tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": true, "model_max_length": 512}

train_args.json

@@ -0,0 +1,25 @@
+{
+  "model": "bert-base-uncased",
+  "dataset": "yelp_polarity",
+  "dataset_train_split": "train",
+  "dataset_dev_split": "test",
+  "tb_writer_step": 1000,
+  "checkpoint_steps": -1,
+  "checkpoint_every_epoch": false,
+  "num_train_epochs": 5,
+  "early_stopping_epochs": -1,
+  "batch_size": 16,
+  "max_length": 256,
+  "learning_rate": 5e-05,
+  "grad_accum_steps": 1,
+  "warmup_proportion": 0.1,
+  "config_name": "config.json",
+  "weights_name": "pytorch_model.bin",
+  "enable_wandb": false,
+  "output_dir": "/p/qdata/jm8wx/research/text_attacks/textattack/outputs/training/bert-base-uncased-yelp_polarity-2020-07-08-10:42/",
+  "num_labels": 2,
+  "do_regression": false,
+  "best_eval_score": 0.9699473684210527,
+  "best_eval_score_epoch": 4,
+  "epochs_since_best_eval_score": 0
+}