Add codes for Phase 1 with data augmentation

JingqingZ · Mar 22, 2019 · c087496 · c087496
1 parent 4be2a49
commit c087496
Show file tree

Hide file tree

Showing 4 changed files with 842 additions and 3 deletions.
diff --git a/README.md b/README.md
@@ -115,6 +115,28 @@ python3 train_reject.py \
         --train 1
 ```
 
+- With data augmentation: an example
+```bash
+python3 train_reject_augmented.py \
+        --data dbpedia \
+        --unseen 0.5 \
+        --model vw \
+        --nepoch 3 \
+        --rgidx 1 \
+        --naug 100 \
+        --train 1
+```
+
+The arguments of the command represent
+* `data`: Dataset, either `dbpedia` or `20news`.
+* `unseen`: Rate of unseen classes, either `0.25` or `0.5`.
+* `model`: The model to be trained. This argument can only be
+    * `vw`: the inputs are embedding of words (from text)
+* `nepoch`: The number of epochs for training
+* `train`: In Phase 1, this argument does not affect the program. The program will run training and testing together.
+* `rgidx`: Optional, Random group starting index: e.g. if 5, the training will start from the 5th random group, by default `1`. This argument is used when the program is accidentally interrupted.
+* `naug`: The number of augmented data per unseen class
+
 
 ### How to train / test the traditional classifier in Phase 2
 
@@ -134,7 +156,7 @@ The arguments of the command represent
 * `model`: The model to be trained. This argument can only be
     * `vw`: the inputs are embedding of words (from text)
 * `sepoch`: Repeat training of each epoch for several times. The ratio of positive/negative samples and learning rate will keep consistent in one epoch no matter how many times the epoch is repeated.
-* `train`: In Phase 1, this argument does not affect the program. The program will run training and testing together.
+* `train`: For the traditional classifier, this argument does not affect the program. The program will run training and testing together.
 * `rgidx`: Optional, Random group starting index: e.g. if 5, the training will start from the 5th random group, by default `1`. This argument is used when the program is accidentally interrupted.
 * `gpu`: Optional, GPU occupation percentage, by default `1.0`, which means full occupation of available GPUs.
 * `baseepoch`: Optional, you may want to specify which epoch to test.

diff --git a/src_reject/config.py b/src_reject/config.py
@@ -17,6 +17,7 @@
 parser.add_argument("--fulltest", type=int, required=False, help="full test or not")
 parser.add_argument("--threshold", type=float, required=False, help="threshold for seen")
 parser.add_argument("--nott", type=int, required=False, help="no. of original texts to be translated")
+parser.add_argument("--naug", type=int, default = 100, required=False, help="no. of augmented data per unseen class")
 args = parser.parse_args()
 print(args)
 
@@ -171,6 +172,7 @@
 zhang15_dbpedia_dir = zhang15_dir + "dbpedia_csv/"
 
 zhang15_dbpedia_full_data_path = zhang15_dbpedia_dir + "full.csv"
+zhang15_dbpedia_full_augmented_path = zhang15_dbpedia_dir + "full_augmented.csv"
 
 zhang15_dbpedia_train_path = zhang15_dbpedia_dir + "train.csv"
 zhang15_dbpedia_train_processed_path = zhang15_dbpedia_dir + "processed_train_text.pkl"

diff --git a/src_reject/train_reject.py b/src_reject/train_reject.py
@@ -557,9 +557,9 @@ def print_summary_of_all_iterations(iteration_statistics):
     print(len(train_class_list), len(train_text_seqs), len(test_class_list), len(test_text_seqs))
     iteration_statistics = []
     num_epoch = config.args.nepoch
-
+    unseen_percentage = config.unseen_rate
+
     for i in range(config.args.rgidx-1, 10):
-        unseen_percentage = config.unseen_rate
         seen_classes, unseen_classes = random_set[i]['seen'], random_set[i]['unseen']
         print(seen_classes, unseen_classes)
         stat_list = []