.

.travis.yml

@@ -18,12 +18,9 @@ before_install:
   - source activate test-environment
   # use requirements.txt for dependencies
   - pip install -r requirements.txt
-  # dependencies for im2text
-  - pip install Pillow
-  - conda install torchvision==0.1.8
-  # dependencies for speech2text
+  # optional dependencies for im2text and speech2text
   - sudo apt-get install -y sox libsox-dev libsox-fmt-all
-  - pip install librosa git+https://github.com/pytorch/audio
+  - pip install -r requirements.opt.txt
 
 
 install:
@@ -50,6 +47,9 @@ script:
   - head /tmp/speech/src-val.txt > /tmp/speech/src-val-head.txt; head /tmp/speech/tgt-val.txt > /tmp/speech/tgt-val-head.txt; python translate.py -data_type audio -src_dir /tmp/speech/an4_dataset -model /tmp/test_model_speech.pt -src /tmp/speech/src-val-head.txt -tgt /tmp/speech/tgt-val-head.txt -verbose -out /tmp/speech/trans; diff /tmp/speech/tgt-val-head.txt /tmp/speech/trans
   # test nmt preprocessing and training
   - head data/src-val.txt > /tmp/src-val.txt; head data/tgt-val.txt > /tmp/tgt-val.txt; python preprocess.py -train_src /tmp/src-val.txt -train_tgt /tmp/tgt-val.txt -valid_src /tmp/src-val.txt -valid_tgt /tmp/tgt-val.txt -save_data /tmp/q -src_vocab_size 1000 -tgt_vocab_size 1000; python train.py -data /tmp/q -rnn_size 2 -batch_size 10 -word_vec_size 5 -report_every 5 -rnn_size 10 -epochs 1
+  # test nmt preprocessing w/ sharding and training w/copy
+    head data/src-val.txt > /tmp/src-val.txt; head data/tgt-val.txt > /tmp/tgt-val.txt; python preprocess.py -train_src /tmp/src-val.txt -train_tgt /tmp/tgt-val.txt -valid_src /tmp/src-val.txt -valid_tgt /tmp/tgt-val.txt -max_shard_size 1 -dynamic_dict -save_data /tmp/q -src_vocab_size 1000 -tgt_vocab_size 1000; python train.py -data /tmp/q -rnn_size 2 -batch_size 10 -word_vec_size 5 -report_every 5 -rnn_size 10 -copy_attn -epochs 1
+
   # test im2text preprocessing and training
   - head /tmp/im2text/src-val.txt > /tmp/im2text/src-val-head.txt; head /tmp/im2text/tgt-val.txt > /tmp/im2text/tgt-val-head.txt; python preprocess.py -data_type img -src_dir /tmp/im2text/images -train_src /tmp/im2text/src-val-head.txt -train_tgt /tmp/im2text/tgt-val-head.txt -valid_src /tmp/im2text/src-val-head.txt -valid_tgt /tmp/im2text/tgt-val-head.txt -save_data /tmp/im2text/q; python train.py -model_type img -data /tmp/im2text/q -rnn_size 2 -batch_size 10 -word_vec_size 5 -report_every 5 -rnn_size 10 -epochs 1
   # test speech2text preprocessing and training

requirements.opt.txt

@@ -1,5 +1,5 @@
 cffi
-torchvision
+torchvision==0.1.8
 librosa
 Pillow
 git+https://github.com/pytorch/audio

test/pull_request_chk.sh

@@ -174,6 +174,27 @@ diff ${DATA_DIR}/morph/tgt.valid /tmp/trans
 echo "Succeeded" | tee -a ${LOG_FILE}
 
 
+# NMT Preprocess w/sharding + Train w/copy
+echo -n "[+] Doing NMT {preprocess w/sharding + train w/copy} test..."
+head ${DATA_DIR}/src-val.txt > /tmp/src-val.txt
+head ${DATA_DIR}/tgt-val.txt > /tmp/tgt-val.txt
+rm -rf /tmp/q*pt
+${PYTHON} preprocess.py -train_src /tmp/src-val.txt \
+		     -train_tgt /tmp/tgt-val.txt \
+		     -valid_src /tmp/src-val.txt \
+		     -valid_tgt /tmp/tgt-val.txt \
+		     -save_data /tmp/q           \
+		     -src_vocab_size 1000        \
+		     -tgt_vocab_size 1000        \
+		     -max_shard_size 1           \
+             -dynamic_dict               >> ${LOG_FILE} 2>&1
+${PYTHON} train.py -data /tmp/q -rnn_size 2 -batch_size 10 \
+		-word_vec_size 5 -report_every 5        \
+		-rnn_size 10 -epochs 1 -copy_attn       >> ${LOG_FILE} 2>&1
+[ "$?" -eq 0 ] || error_exit
+echo "Succeeded" | tee -a ${LOG_FILE}
+
+
 echo -n "[+] Doing im2text {preprocess + train} test..."
 head /tmp/im2text/src-val.txt > /tmp/im2text/src-val-head.txt
 head /tmp/im2text/tgt-val.txt > /tmp/im2text/tgt-val-head.txt