summary all

summary_all_scandflan_base.sh

@@ -0,0 +1,14 @@
+PROJECT_DIR=${HOME}"/models/t5-summary"
+export PYTHONPATH=${PROJECT_DIR}
+INITIAL_CHECKPOINT_PATH=\"gs://nb-t5x-us-central2/exp5-t5-base-ul2-scandvoc-full/checkpoint_1500000\"
+TRAIN_STEPS=1550000
+
+
+
+python3 ../../t5x/t5x/train.py \
+  --gin_search_paths="./" \
+  --gin.TRAIN_STEPS=${TRAIN_STEPS} \
+  --gin_file="finetune_summary_scandflan_base.gin" \
+  --gin.MIXTURE_OR_TASK_NAME=\"summary_all_scand\" \
+  --gin.MODEL_DIR=\"gs://nb-t5x-us-central2/finetuned/summary_scandflan_all_base\" \
+  --gin.INITIAL_CHECKPOINT_PATH=${INITIAL_CHECKPOINT_PATH} \

tasks.py

@@ -247,4 +247,20 @@ seqio.TaskRegistry.add(
     metric_fns=[metrics.accuracy,my_metrics.f1_macro,metrics.bleu,metrics.rouge],
     output_features=MT5_OUTPUT_FEATURES,
 )   
+seqio.TaskRegistry.add(
+    "summary_all_scand",
+    source=seqio.TextLineDataSource(
+        split_to_filepattern=tsv_summary_all_path,
+        #num_input_examples=num_nq_examples
+        ),
+    preprocessors=[
+      functools.partial(
+          t5.data.preprocessors.parse_tsv,
+          field_names=["source","target"]),
+      categorise_preprocessor,
+      seqio.preprocessors.tokenize_and_append_eos,
+    ],
+    metric_fns=[metrics.accuracy,my_metrics.f1_macro,metrics.bleu,metrics.rouge],
+    output_features=SCAND_OUTPUT_FEATURES,
+)