model upload

cnn-chinanews-chinese/README.md

@@ -0,0 +1,10 @@
+## TextAttack Model Card
+
+            This  model was fine-tuned using TextAttack. The model was fine-tuned
+            for 10 epochs with a batch size of 16,
+             and an initial learning rate of 0.0002.
+            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 0.8842, as measured by the
+            eval set accuracy, found after 2 epochs.
+
+            For more information, check out [TextAttack on Github](https://github.com/QData/TextAttack).

cnn-chinanews-chinese/config.json

@@ -0,0 +1 @@
+{"architectures": "WordCNNForClassification", "hidden_size": 150, "dropout": 0.3, "num_labels": 2, "max_seq_length": 128, "model_path": null, "emb_layer_trainable": true}

cnn-chinanews-chinese/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17c320a4c4daeb8867144410c627d33f15335e506ade22331addce36278fc95a
+size 765335626

cnn-chinanews-chinese/train_log.txt

@@ -0,0 +1,63 @@
+Writing logs to ./outputs/2024-03-04-17-36-59-095657/train_log.txt.
+Wrote original training args to ./outputs/2024-03-04-17-36-59-095657/training_args.json.
+***** Running training *****
+  Num examples = 50000
+  Num epochs = 10
+  Num clean epochs = 10
+  Instantaneous batch size per device = 16
+  Total train batch size (w. parallel, distributed & accumulation) = 16
+  Gradient accumulation steps = 1
+  Total optimization steps = 31250
+==========================================================
+Epoch 1
+Running clean epoch 1/10
+Train accuracy: 78.91%
+Eval accuracy: 87.58%
+Best score found. Saved model to ./outputs/2024-03-04-17-36-59-095657/best_model/
+==========================================================
+Epoch 2
+Running clean epoch 2/10
+Train accuracy: 90.36%
+Eval accuracy: 88.42%
+Best score found. Saved model to ./outputs/2024-03-04-17-36-59-095657/best_model/
+==========================================================
+Epoch 3
+Running clean epoch 3/10
+Train accuracy: 94.05%
+Eval accuracy: 88.36%
+==========================================================
+Epoch 4
+Running clean epoch 4/10
+Train accuracy: 96.72%
+Eval accuracy: 87.84%
+==========================================================
+Epoch 5
+Running clean epoch 5/10
+Train accuracy: 98.54%
+Eval accuracy: 87.60%
+==========================================================
+Epoch 6
+Running clean epoch 6/10
+Train accuracy: 99.33%
+Eval accuracy: 86.59%
+==========================================================
+Epoch 7
+Running clean epoch 7/10
+Train accuracy: 99.75%
+Eval accuracy: 86.52%
+==========================================================
+Epoch 8
+Running clean epoch 8/10
+Train accuracy: 99.91%
+Eval accuracy: 86.28%
+==========================================================
+Epoch 9
+Running clean epoch 9/10
+Train accuracy: 99.96%
+Eval accuracy: 86.42%
+==========================================================
+Epoch 10
+Running clean epoch 10/10
+Train accuracy: 99.96%
+Eval accuracy: 86.06%
+Wrote README to ./outputs/2024-03-04-17-36-59-095657/README.md.

cnn-chinanews-chinese/training_args.json

@@ -0,0 +1 @@
+{"num_epochs": 10, "num_clean_epochs": 1, "attack_epoch_interval": 1, "early_stopping_epochs": null, "learning_rate": 0.0002, "num_warmup_steps": 500, "weight_decay": 0.01, "per_device_train_batch_size": 16, "per_device_eval_batch_size": 32, "gradient_accumulation_steps": 1, "random_seed": 718, "parallel": false, "load_best_model_at_end": false, "alpha": 1.0, "num_train_adv_examples": -1, "query_budget_train": null, "attack_num_workers_per_device": 1, "output_dir": "./outputs/2024-03-04-17-36-59-095657", "checkpoint_interval_steps": null, "checkpoint_interval_epochs": null, "save_last": true, "log_to_tb": false, "tb_log_dir": null, "log_to_wandb": false, "wandb_project": "textattack", "logging_interval_step": 1}

cnn-chnsenticorp-chinese/README.md

@@ -0,0 +1,10 @@
+## TextAttack Model Card
+
+            This  model was fine-tuned using TextAttack. The model was fine-tuned
+            for 10 epochs with a batch size of 16,
+             and an initial learning rate of 0.0002.
+            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 0.8966666666666666, as measured by the
+            eval set accuracy, found after 3 epochs.
+
+            For more information, check out [TextAttack on Github](https://github.com/QData/TextAttack).

cnn-chnsenticorp-chinese/config.json

@@ -0,0 +1 @@
+{"architectures": "WordCNNForClassification", "hidden_size": 150, "dropout": 0.3, "num_labels": 2, "max_seq_length": 128, "model_path": null, "emb_layer_trainable": true}

cnn-chnsenticorp-chinese/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9794c7a2ee57d0b61814f9a2d8e5f8455d2cd99ebf965aec6f1ff12ae507a46
+size 765326666

cnn-chnsenticorp-chinese/train_log.txt

@@ -0,0 +1,64 @@
+Writing logs to ./outputs/2024-03-02-22-47-16-038201/train_log.txt.
+Wrote original training args to ./outputs/2024-03-02-22-47-16-038201/training_args.json.
+***** Running training *****
+  Num examples = 9600
+  Num epochs = 10
+  Num clean epochs = 10
+  Instantaneous batch size per device = 16
+  Total train batch size (w. parallel, distributed & accumulation) = 16
+  Gradient accumulation steps = 1
+  Total optimization steps = 6000
+==========================================================
+Epoch 1
+Running clean epoch 1/10
+Train accuracy: 73.18%
+Eval accuracy: 82.58%
+Best score found. Saved model to ./outputs/2024-03-02-22-47-16-038201/best_model/
+==========================================================
+Epoch 2
+Running clean epoch 2/10
+Train accuracy: 88.45%
+Eval accuracy: 86.92%
+Best score found. Saved model to ./outputs/2024-03-02-22-47-16-038201/best_model/
+==========================================================
+Epoch 3
+Running clean epoch 3/10
+Train accuracy: 93.04%
+Eval accuracy: 89.67%
+Best score found. Saved model to ./outputs/2024-03-02-22-47-16-038201/best_model/
+==========================================================
+Epoch 4
+Running clean epoch 4/10
+Train accuracy: 96.05%
+Eval accuracy: 88.75%
+==========================================================
+Epoch 5
+Running clean epoch 5/10
+Train accuracy: 97.74%
+Eval accuracy: 89.50%
+==========================================================
+Epoch 6
+Running clean epoch 6/10
+Train accuracy: 98.78%
+Eval accuracy: 89.50%
+==========================================================
+Epoch 7
+Running clean epoch 7/10
+Train accuracy: 99.17%
+Eval accuracy: 89.33%
+==========================================================
+Epoch 8
+Running clean epoch 8/10
+Train accuracy: 99.22%
+Eval accuracy: 88.92%
+==========================================================
+Epoch 9
+Running clean epoch 9/10
+Train accuracy: 99.43%
+Eval accuracy: 89.17%
+==========================================================
+Epoch 10
+Running clean epoch 10/10
+Train accuracy: 99.41%
+Eval accuracy: 88.17%
+Wrote README to ./outputs/2024-03-02-22-47-16-038201/README.md.

cnn-chnsenticorp-chinese/training_args.json

@@ -0,0 +1 @@
+{"num_epochs": 10, "num_clean_epochs": 1, "attack_epoch_interval": 1, "early_stopping_epochs": null, "learning_rate": 0.0002, "num_warmup_steps": 500, "weight_decay": 0.01, "per_device_train_batch_size": 16, "per_device_eval_batch_size": 32, "gradient_accumulation_steps": 1, "random_seed": 718, "parallel": false, "load_best_model_at_end": false, "alpha": 1.0, "num_train_adv_examples": -1, "query_budget_train": null, "attack_num_workers_per_device": 1, "output_dir": "./outputs/2024-03-02-22-47-16-038201", "checkpoint_interval_steps": null, "checkpoint_interval_epochs": null, "save_last": true, "log_to_tb": false, "tb_log_dir": null, "log_to_wandb": false, "wandb_project": "textattack", "logging_interval_step": 1}

lstm-chinanews-chinese/README.md

@@ -0,0 +1,10 @@
+## TextAttack Model Card
+
+            This  model was fine-tuned using TextAttack. The model was fine-tuned
+            for 10 epochs with a batch size of 16,
+             and an initial learning rate of 0.0002.
+            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 0.8767, as measured by the
+            eval set accuracy, found after 3 epochs.
+
+            For more information, check out [TextAttack on Github](https://github.com/QData/TextAttack).

lstm-chinanews-chinese/config.json

@@ -0,0 +1 @@
+{"architectures": "LSTMForClassification", "hidden_size": 150, "depth": 1, "dropout": 0.3, "num_labels": 2, "max_seq_length": 128, "model_path": null, "emb_layer_trainable": true}

lstm-chinanews-chinese/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:09d1e39a6965e05c1dcb27dc2ba86a7b06d686812ca359d4fc440282535be04a
+size 764070671

lstm-chinanews-chinese/train_log.txt

@@ -0,0 +1,64 @@
+Writing logs to ./outputs/2024-03-04-17-14-45-472632/train_log.txt.
+Wrote original training args to ./outputs/2024-03-04-17-14-45-472632/training_args.json.
+***** Running training *****
+  Num examples = 50000
+  Num epochs = 10
+  Num clean epochs = 10
+  Instantaneous batch size per device = 16
+  Total train batch size (w. parallel, distributed & accumulation) = 16
+  Gradient accumulation steps = 1
+  Total optimization steps = 31250
+==========================================================
+Epoch 1
+Running clean epoch 1/10
+Train accuracy: 73.86%
+Eval accuracy: 85.25%
+Best score found. Saved model to ./outputs/2024-03-04-17-14-45-472632/best_model/
+==========================================================
+Epoch 2
+Running clean epoch 2/10
+Train accuracy: 88.42%
+Eval accuracy: 87.05%
+Best score found. Saved model to ./outputs/2024-03-04-17-14-45-472632/best_model/
+==========================================================
+Epoch 3
+Running clean epoch 3/10
+Train accuracy: 92.21%
+Eval accuracy: 87.67%
+Best score found. Saved model to ./outputs/2024-03-04-17-14-45-472632/best_model/
+==========================================================
+Epoch 4
+Running clean epoch 4/10
+Train accuracy: 94.91%
+Eval accuracy: 87.34%
+==========================================================
+Epoch 5
+Running clean epoch 5/10
+Train accuracy: 97.15%
+Eval accuracy: 87.03%
+==========================================================
+Epoch 6
+Running clean epoch 6/10
+Train accuracy: 98.45%
+Eval accuracy: 86.12%
+==========================================================
+Epoch 7
+Running clean epoch 7/10
+Train accuracy: 99.25%
+Eval accuracy: 85.90%
+==========================================================
+Epoch 8
+Running clean epoch 8/10
+Train accuracy: 99.60%
+Eval accuracy: 85.51%
+==========================================================
+Epoch 9
+Running clean epoch 9/10
+Train accuracy: 99.82%
+Eval accuracy: 85.36%
+==========================================================
+Epoch 10
+Running clean epoch 10/10
+Train accuracy: 99.91%
+Eval accuracy: 84.99%
+Wrote README to ./outputs/2024-03-04-17-14-45-472632/README.md.

lstm-chinanews-chinese/training_args.json

@@ -0,0 +1 @@
+{"num_epochs": 10, "num_clean_epochs": 1, "attack_epoch_interval": 1, "early_stopping_epochs": null, "learning_rate": 0.0002, "num_warmup_steps": 500, "weight_decay": 0.01, "per_device_train_batch_size": 16, "per_device_eval_batch_size": 32, "gradient_accumulation_steps": 1, "random_seed": 718, "parallel": false, "load_best_model_at_end": false, "alpha": 1.0, "num_train_adv_examples": -1, "query_budget_train": null, "attack_num_workers_per_device": 1, "output_dir": "./outputs/2024-03-04-17-14-45-472632", "checkpoint_interval_steps": null, "checkpoint_interval_epochs": null, "save_last": true, "log_to_tb": false, "tb_log_dir": null, "log_to_wandb": false, "wandb_project": "textattack", "logging_interval_step": 1}

lstm-chnsenticorp-chinese/README.md

@@ -0,0 +1,10 @@
+## TextAttack Model Card
+
+            This  model was fine-tuned using TextAttack. The model was fine-tuned
+            for 10 epochs with a batch size of 16,
+             and an initial learning rate of 0.0002.
+            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 0.8983333333333333, as measured by the
+            eval set accuracy, found after 7 epochs.
+
+            For more information, check out [TextAttack on Github](https://github.com/QData/TextAttack).

lstm-chnsenticorp-chinese/config.json

@@ -0,0 +1 @@
+{"architectures": "LSTMForClassification", "hidden_size": 150, "depth": 1, "dropout": 0.3, "num_labels": 2, "max_seq_length": 128, "model_path": null, "emb_layer_trainable": true}

lstm-chnsenticorp-chinese/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9db315e2c8c53b0a5f53bad2628d634641507b2fa2ece02942c1f5baa2ce36dc
+size 764067663

lstm-chnsenticorp-chinese/train_log.txt

@@ -0,0 +1,67 @@
+Writing logs to ./outputs/2024-03-02-22-34-05-343811/train_log.txt.
+Wrote original training args to ./outputs/2024-03-02-22-34-05-343811/training_args.json.
+***** Running training *****
+  Num examples = 9600
+  Num epochs = 10
+  Num clean epochs = 10
+  Instantaneous batch size per device = 16
+  Total train batch size (w. parallel, distributed & accumulation) = 16
+  Gradient accumulation steps = 1
+  Total optimization steps = 6000
+==========================================================
+Epoch 1
+Running clean epoch 1/10
+Train accuracy: 69.30%
+Eval accuracy: 75.33%
+Best score found. Saved model to ./outputs/2024-03-02-22-34-05-343811/best_model/
+==========================================================
+Epoch 2
+Running clean epoch 2/10
+Train accuracy: 84.70%
+Eval accuracy: 85.50%
+Best score found. Saved model to ./outputs/2024-03-02-22-34-05-343811/best_model/
+==========================================================
+Epoch 3
+Running clean epoch 3/10
+Train accuracy: 90.29%
+Eval accuracy: 87.92%
+Best score found. Saved model to ./outputs/2024-03-02-22-34-05-343811/best_model/
+==========================================================
+Epoch 4
+Running clean epoch 4/10
+Train accuracy: 93.79%
+Eval accuracy: 89.00%
+Best score found. Saved model to ./outputs/2024-03-02-22-34-05-343811/best_model/
+==========================================================
+Epoch 5
+Running clean epoch 5/10
+Train accuracy: 96.18%
+Eval accuracy: 88.67%
+==========================================================
+Epoch 6
+Running clean epoch 6/10
+Train accuracy: 97.52%
+Eval accuracy: 89.50%
+Best score found. Saved model to ./outputs/2024-03-02-22-34-05-343811/best_model/
+==========================================================
+Epoch 7
+Running clean epoch 7/10
+Train accuracy: 98.43%
+Eval accuracy: 89.83%
+Best score found. Saved model to ./outputs/2024-03-02-22-34-05-343811/best_model/
+==========================================================
+Epoch 8
+Running clean epoch 8/10
+Train accuracy: 98.95%
+Eval accuracy: 88.92%
+==========================================================
+Epoch 9
+Running clean epoch 9/10
+Train accuracy: 99.25%
+Eval accuracy: 89.17%
+==========================================================
+Epoch 10
+Running clean epoch 10/10
+Train accuracy: 99.39%
+Eval accuracy: 88.42%
+Wrote README to ./outputs/2024-03-02-22-34-05-343811/README.md.

lstm-chnsenticorp-chinese/training_args.json

@@ -0,0 +1 @@
+{"num_epochs": 10, "num_clean_epochs": 1, "attack_epoch_interval": 1, "early_stopping_epochs": null, "learning_rate": 0.0002, "num_warmup_steps": 500, "weight_decay": 0.01, "per_device_train_batch_size": 16, "per_device_eval_batch_size": 32, "gradient_accumulation_steps": 1, "random_seed": 718, "parallel": false, "load_best_model_at_end": false, "alpha": 1.0, "num_train_adv_examples": -1, "query_budget_train": null, "attack_num_workers_per_device": 1, "output_dir": "./outputs/2024-03-02-22-34-05-343811", "checkpoint_interval_steps": null, "checkpoint_interval_epochs": null, "save_last": true, "log_to_tb": false, "tb_log_dir": null, "log_to_wandb": false, "wandb_project": "textattack", "logging_interval_step": 1}