From 731a92b301a1f773753f87b123be437eaa951245 Mon Sep 17 00:00:00 2001
From: zqwerty <zhuq96@hotmail.com>
Date: Sun, 10 Apr 2022 13:32:01 +0800
Subject: [PATCH] add evaluate script for t5/(nlu|dst)/*.sh
---
convlab2/base_models/t5/dst/run_multiwoz21.sh | 2 ++
convlab2/base_models/t5/nlu/run_multiwoz21_user.sh | 2 ++
convlab2/base_models/t5/nlu/run_multiwoz21_user_context3.sh | 2 ++
convlab2/base_models/t5/nlu/run_tm1_user.sh | 2 ++
convlab2/base_models/t5/nlu/run_tm1_user_context3.sh | 2 ++
convlab2/base_models/t5/nlu/run_tm2_user.sh | 2 ++
convlab2/base_models/t5/nlu/run_tm2_user_context3.sh | 2 ++
convlab2/base_models/t5/nlu/run_tm3_user.sh | 2 ++
convlab2/base_models/t5/nlu/run_tm3_user_context3.sh | 2 ++
9 files changed, 18 insertions(+)
diff --git a/convlab2/base_models/t5/dst/run_multiwoz21.sh b/convlab2/base_models/t5/dst/run_multiwoz21.sh
index e7573e95..6d42380b 100644
--- a/convlab2/base_models/t5/dst/run_multiwoz21.sh
+++ b/convlab2/base_models/t5/dst/run_multiwoz21.sh
@@ -80,3 +80,5 @@ python -m torch.distributed.launch \
--per_device_eval_batch_size ${per_device_eval_batch_size} \
python merge_predict_res.py -d ${dataset_name} -s ${speaker} -c ${context_window_size} -p ${output_dir}/generated_predictions.json
+
+python ../../../dst/evaluate_unified_datasets.py -p ${output_dir}/predictions.json
diff --git a/convlab2/base_models/t5/nlu/run_multiwoz21_user.sh b/convlab2/base_models/t5/nlu/run_multiwoz21_user.sh
index 4080a092..a9e9d6c5 100644
--- a/convlab2/base_models/t5/nlu/run_multiwoz21_user.sh
+++ b/convlab2/base_models/t5/nlu/run_multiwoz21_user.sh
@@ -64,3 +64,5 @@ python -m torch.distributed.launch \
--per_device_eval_batch_size ${per_device_eval_batch_size} \
python merge_predict_res.py -d ${dataset_name} -s ${speaker} -c ${context_window_size} -p ${output_dir}/generated_predictions.json
+
+python ../../../nlu/evaluate_unified_datasets.py -p ${output_dir}/predictions.json
diff --git a/convlab2/base_models/t5/nlu/run_multiwoz21_user_context3.sh b/convlab2/base_models/t5/nlu/run_multiwoz21_user_context3.sh
index a8cbc9bd..e90e71d4 100644
--- a/convlab2/base_models/t5/nlu/run_multiwoz21_user_context3.sh
+++ b/convlab2/base_models/t5/nlu/run_multiwoz21_user_context3.sh
@@ -64,3 +64,5 @@ python -m torch.distributed.launch \
--per_device_eval_batch_size ${per_device_eval_batch_size} \
python merge_predict_res.py -d ${dataset_name} -s ${speaker} -c ${context_window_size} -p ${output_dir}/generated_predictions.json
+
+python ../../../nlu/evaluate_unified_datasets.py -p ${output_dir}/predictions.json
diff --git a/convlab2/base_models/t5/nlu/run_tm1_user.sh b/convlab2/base_models/t5/nlu/run_tm1_user.sh
index 9faedd7f..5372441a 100644
--- a/convlab2/base_models/t5/nlu/run_tm1_user.sh
+++ b/convlab2/base_models/t5/nlu/run_tm1_user.sh
@@ -64,3 +64,5 @@ python -m torch.distributed.launch \
--per_device_eval_batch_size ${per_device_eval_batch_size} \
python merge_predict_res.py -d ${dataset_name} -s ${speaker} -c ${context_window_size} -p ${output_dir}/generated_predictions.json
+
+python ../../../nlu/evaluate_unified_datasets.py -p ${output_dir}/predictions.json
diff --git a/convlab2/base_models/t5/nlu/run_tm1_user_context3.sh b/convlab2/base_models/t5/nlu/run_tm1_user_context3.sh
index bb6b55fe..65482a1f 100644
--- a/convlab2/base_models/t5/nlu/run_tm1_user_context3.sh
+++ b/convlab2/base_models/t5/nlu/run_tm1_user_context3.sh
@@ -64,3 +64,5 @@ python -m torch.distributed.launch \
--per_device_eval_batch_size ${per_device_eval_batch_size} \
python merge_predict_res.py -d ${dataset_name} -s ${speaker} -c ${context_window_size} -p ${output_dir}/generated_predictions.json
+
+python ../../../nlu/evaluate_unified_datasets.py -p ${output_dir}/predictions.json
diff --git a/convlab2/base_models/t5/nlu/run_tm2_user.sh b/convlab2/base_models/t5/nlu/run_tm2_user.sh
index 728a8a94..84dc8b71 100644
--- a/convlab2/base_models/t5/nlu/run_tm2_user.sh
+++ b/convlab2/base_models/t5/nlu/run_tm2_user.sh
@@ -64,3 +64,5 @@ python -m torch.distributed.launch \
--per_device_eval_batch_size ${per_device_eval_batch_size} \
python merge_predict_res.py -d ${dataset_name} -s ${speaker} -c ${context_window_size} -p ${output_dir}/generated_predictions.json
+
+python ../../../nlu/evaluate_unified_datasets.py -p ${output_dir}/predictions.json
diff --git a/convlab2/base_models/t5/nlu/run_tm2_user_context3.sh b/convlab2/base_models/t5/nlu/run_tm2_user_context3.sh
index 8ebb102d..abca0a60 100644
--- a/convlab2/base_models/t5/nlu/run_tm2_user_context3.sh
+++ b/convlab2/base_models/t5/nlu/run_tm2_user_context3.sh
@@ -64,3 +64,5 @@ python -m torch.distributed.launch \
--per_device_eval_batch_size ${per_device_eval_batch_size} \
python merge_predict_res.py -d ${dataset_name} -s ${speaker} -c ${context_window_size} -p ${output_dir}/generated_predictions.json
+
+python ../../../nlu/evaluate_unified_datasets.py -p ${output_dir}/predictions.json
diff --git a/convlab2/base_models/t5/nlu/run_tm3_user.sh b/convlab2/base_models/t5/nlu/run_tm3_user.sh
index 0d775f7a..689a626c 100644
--- a/convlab2/base_models/t5/nlu/run_tm3_user.sh
+++ b/convlab2/base_models/t5/nlu/run_tm3_user.sh
@@ -64,3 +64,5 @@ python -m torch.distributed.launch \
--per_device_eval_batch_size ${per_device_eval_batch_size} \
python merge_predict_res.py -d ${dataset_name} -s ${speaker} -c ${context_window_size} -p ${output_dir}/generated_predictions.json
+
+python ../../../nlu/evaluate_unified_datasets.py -p ${output_dir}/predictions.json
diff --git a/convlab2/base_models/t5/nlu/run_tm3_user_context3.sh b/convlab2/base_models/t5/nlu/run_tm3_user_context3.sh
index c8a4a9f6..e2ded66d 100644
--- a/convlab2/base_models/t5/nlu/run_tm3_user_context3.sh
+++ b/convlab2/base_models/t5/nlu/run_tm3_user_context3.sh
@@ -64,3 +64,5 @@ python -m torch.distributed.launch \
--per_device_eval_batch_size ${per_device_eval_batch_size} \
python merge_predict_res.py -d ${dataset_name} -s ${speaker} -c ${context_window_size} -p ${output_dir}/generated_predictions.json
+
+python ../../../nlu/evaluate_unified_datasets.py -p ${output_dir}/predictions.json
--
GitLab