From 1f7480a9cb9b50b08938948989ed865307f45c49 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Anast=C3=A1sis?= Date: Thu, 6 Jul 2023 02:35:22 +0400 Subject: [PATCH 01/28] first batch of readmes to the key services, TBC --- annotators/combined_classification/README.md | 1 + .../README.md | 35 +++++++++++++++++++ .../README.md | 2 +- 3 files changed, 37 insertions(+), 1 deletion(-) diff --git a/annotators/combined_classification/README.md b/annotators/combined_classification/README.md index 2b76ce1368..f582b8a095 100644 --- a/annotators/combined_classification/README.md +++ b/annotators/combined_classification/README.md @@ -30,3 +30,4 @@ text here if i/o specified ## Dependencies +For more details please refer to [this script](server.py) \ No newline at end of file diff --git a/annotators/combined_classification_lightweight/README.md b/annotators/combined_classification_lightweight/README.md index 08aea83673..ea11ff5e45 100644 --- a/annotators/combined_classification_lightweight/README.md +++ b/annotators/combined_classification_lightweight/README.md @@ -25,6 +25,41 @@ The models (multitask and comparative single task) were trained with initial lea This model(with a huawei-noah/TinyBERT_General_4L_312D backbone) on a CPU-only inference takes 42% less time than combined_classification, while using only ~1.5 Gb of the CPU instead of the 2909 Mb for combined_classification. The average accuracy and average F1 at the same time are for this model only ~1.5% lower than for the combined_classification, and this dropdown is consistent for all tasks. +# Component Name + +## Description +component_type: multitask model, NN-based +is_customizable: true + +This model is based on the transformer-agnostic multitask neural architecture. It can solve several tasks similtaneously, almost as good as single-task models. + +The models were trained on the following datasets: + +**Factoid classification** : For the Factoid task, we used the same Yahoo ConversVsInfo dataset that was used to train the Dream socialbot in Alexa Prize . Note that the valid set in this task was equal to the test set. + +**Midas classification** : For the Midas task, we used the same Midas classification dataset that was used to train the Dream socialbot in Alexa Prize . Note that the valid set in this task was equal to the test set. + +**Emotion classification** :For the Emotion classification task, we used the emo\_go\_emotions dataset, with all the 28 classes compressed into the seven basic emotions as in the original paper. Note that these 7 emotions are not exactly the same as the 7 emotions in the original Dream socialbot in Alexa Prize: 1 emotion differs (love VS disgust), so the scores are incomparable with the original model. Note that this task is multiclass. + +**Topic classification**: For the Topic classification task, we used the dataset made by Dilyara Zharikova. The dataset was further filtered and improved for the final model version, to make the model suitable for DREAM. Note that the original topics model doesn’t account for that dataset changes(which were also about class number) and thus its scores are not compatible with the scores we have. + +**Sentiment classification** : For the Sentiment classification task, we used the Dynabench dataset (r1 + r2). + +**Toxic classification** : For the toxic classification task, we used the dataset from kaggle the 7 toxic classes that pose an interest to us. Note that this task is multilabel. + +The model also contains 3 replacement models for Amazon services. + +The models (multitask and comparative single task) were trained with initial learning rate 2e-5(with validation patience 2 it could be dropped 2 times), batch size 32,optimizer adamW(betas (0.9,0.99) and early stop on 3 epochs. The criteria on early stopping was average accuracy for all tasks for multitask models, or the single-task accuracy for singletask models. + +This model(with a distilbert-base-uncased backbone) takes only 2439 Mb for 9 tasks, whereas single-task models with the same backbone for every of these tasks take up almost the same memory(~2437 Mb for every of these 9 tasks). + +## I/O +text here if i/o specified + +## Dependencies + +Configuration settings specified in the .yaml files at [service_configs](service_configs) +Required Python packages specified in [requirements.txt](requirements.txt) and [requirements_load_test](requirements_load_test.txt) ## I/O diff --git a/response_selectors/confidence_based_response_selector/README.md b/response_selectors/confidence_based_response_selector/README.md index 0b97298b99..f2eda5a8fe 100644 --- a/response_selectors/confidence_based_response_selector/README.md +++ b/response_selectors/confidence_based_response_selector/README.md @@ -11,4 +11,4 @@ Parameter `FILTER_BADLISTED_WORDS` defines whether it filers out toxic hypothese ## Dependencies -- none \ No newline at end of file +- none From 871fa3ebd441af3cfad1182a44c9845a5396e896 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Anast=C3=A1sis?= Date: Sun, 9 Jul 2023 03:44:48 +0400 Subject: [PATCH 02/28] readmes for dream multilingual dist --- response_selectors/ranking_based_response_selector/README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/response_selectors/ranking_based_response_selector/README.md b/response_selectors/ranking_based_response_selector/README.md index fa7c57f67e..24af087d0a 100644 --- a/response_selectors/ranking_based_response_selector/README.md +++ b/response_selectors/ranking_based_response_selector/README.md @@ -15,4 +15,4 @@ Parameter `FILTER_TOXIC_OR_BADLISTED` defines whether it filers out toxic hypoth ## Dependencies -- either candidate annotations by `SENTENCE_RANKER_ANNOTATION_NAME` or service `SENTENCE_RANKER_SERVICE_URL`. \ No newline at end of file +- either candidate annotations by `SENTENCE_RANKER_ANNOTATION_NAME` or service `SENTENCE_RANKER_SERVICE_URL`. From cb96cbd35226d818bde8bb3600b7d7fc708afcc2 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Anast=C3=A1sis?= Date: Sun, 9 Jul 2023 03:59:35 +0400 Subject: [PATCH 03/28] main distribution readmes upd, TBC --- .../README.md | 42 +++++++++++++++++++ .../llm_based_response_selector/README.md | 2 +- 2 files changed, 43 insertions(+), 1 deletion(-) diff --git a/response_selectors/convers_evaluation_based_selector/README.md b/response_selectors/convers_evaluation_based_selector/README.md index fb14cb0daa..026e19464d 100644 --- a/response_selectors/convers_evaluation_based_selector/README.md +++ b/response_selectors/convers_evaluation_based_selector/README.md @@ -38,4 +38,46 @@ FALLBACK_FILE: a file name with fallbacks from `dream/common/fallbacks/` ## Dependencies +<<<<<<< HEAD +======= +For this context you may type 0,3 hypots: +``` +Type best hypot num(s), separated by comma: 0,3 +``` + +When dialog ends it saves it into labeled_data folder. + +### Measuring quality + +It takes `--data_dir` as an input (the same as save_dir from data labelling script). +Outputs overall accuracy. + +Example of usage: + +``` +python response_selectors/convers_evaluation_based_selector/measure_quality.py \ + --data_dir response_selectors/convers_evaluation_based_selector/labeled_data/ + +Overall accuracy: 0.5185185185185185 +``` + +### How to run conversation evaluator locally + +`docker-compose -f docker-compose.yml -f dev.yml -f cpu.yml -f one_worker.yml up toxic_classification badlisted_words convers_evaluation_selector` + +Then use `--url`. + +Example of usage with url: + +``` +python response_selectors/convers_evaluation_based_selector/measure_quality.py \ + --data_dir response_selectors/convers_evaluation_based_selector/labeled_data/ \ + --url http://0.0.0.0:8009/respond +``` +## Dependencies + +Configuration settings specified in the .yaml files at [service_configs](service_configs) +Required Python packages specified in [requirements.txt](requirements.txt). +For more details please refer to [this script](server.py) " +>>>>>>> main distribution readmes upd, TBC diff --git a/response_selectors/llm_based_response_selector/README.md b/response_selectors/llm_based_response_selector/README.md index e74d8c8cf9..fcb4024a74 100644 --- a/response_selectors/llm_based_response_selector/README.md +++ b/response_selectors/llm_based_response_selector/README.md @@ -27,4 +27,4 @@ Parameter `FILTER_TOXIC_OR_BADLISTED` defines whether it filers out toxic hypoth ## Dependencies -- generative service `GENERATIVE_SERVICE_URL` \ No newline at end of file +- generative service `GENERATIVE_SERVICE_URL` From cd3fef8b28b37ebad0b36827f8956f69868602b1 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Anast=C3=A1sis?= Date: Thu, 6 Jul 2023 02:35:22 +0400 Subject: [PATCH 04/28] first batch of readmes to the key services, TBC --- annotators/combined_classification/README.md | 7 +++++++ annotators/combined_classification_lightweight/README.md | 7 +++++++ 2 files changed, 14 insertions(+) diff --git a/annotators/combined_classification/README.md b/annotators/combined_classification/README.md index f582b8a095..147f5ac2a5 100644 --- a/annotators/combined_classification/README.md +++ b/annotators/combined_classification/README.md @@ -2,6 +2,13 @@ # Combined_classification ## Description +<<<<<<< HEAD +======= +component_type: multitask model, NN-based +is_customizable: true +CPU memory use: 2909 Mb. + +>>>>>>> first batch of readmes to the key services, TBC This model is based on the transformer-agnostic multitask neural architecture. It can solve several tasks similtaneously, almost as good as single-task models. diff --git a/annotators/combined_classification_lightweight/README.md b/annotators/combined_classification_lightweight/README.md index ea11ff5e45..15f898b7e9 100644 --- a/annotators/combined_classification_lightweight/README.md +++ b/annotators/combined_classification_lightweight/README.md @@ -2,6 +2,12 @@ # Combined_classification_lightweight ## Description +<<<<<<< HEAD +======= +component_type: multitask model, NN-based +is_customizable: true +CPU memory use: 2909 Mb +>>>>>>> first batch of readmes to the key services, TBC This model is based on the transformer-agnostic multitask neural architecture. It can solve several tasks similtaneously, almost as good as single-task models. @@ -66,3 +72,4 @@ Required Python packages specified in [requirements.txt](requirements.txt) and [ ## Dependencies +For more details please refer to [this script](server.py) From c0ed857022066afac9178845a8cafe34711f2e47 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Anast=C3=A1sis?= Date: Sun, 9 Jul 2023 03:44:48 +0400 Subject: [PATCH 05/28] readmes for dream multilingual dist --- annotators/combined_classification/README.md | 7 ------ .../README.md | 6 ----- .../ranking_based_response_selector/README.md | 23 +++++++++++++++++++ 3 files changed, 23 insertions(+), 13 deletions(-) diff --git a/annotators/combined_classification/README.md b/annotators/combined_classification/README.md index 147f5ac2a5..f582b8a095 100644 --- a/annotators/combined_classification/README.md +++ b/annotators/combined_classification/README.md @@ -2,13 +2,6 @@ # Combined_classification ## Description -<<<<<<< HEAD -======= -component_type: multitask model, NN-based -is_customizable: true -CPU memory use: 2909 Mb. - ->>>>>>> first batch of readmes to the key services, TBC This model is based on the transformer-agnostic multitask neural architecture. It can solve several tasks similtaneously, almost as good as single-task models. diff --git a/annotators/combined_classification_lightweight/README.md b/annotators/combined_classification_lightweight/README.md index 15f898b7e9..c1e485b7e1 100644 --- a/annotators/combined_classification_lightweight/README.md +++ b/annotators/combined_classification_lightweight/README.md @@ -2,12 +2,6 @@ # Combined_classification_lightweight ## Description -<<<<<<< HEAD -======= -component_type: multitask model, NN-based -is_customizable: true -CPU memory use: 2909 Mb ->>>>>>> first batch of readmes to the key services, TBC This model is based on the transformer-agnostic multitask neural architecture. It can solve several tasks similtaneously, almost as good as single-task models. diff --git a/response_selectors/ranking_based_response_selector/README.md b/response_selectors/ranking_based_response_selector/README.md index 24af087d0a..f028664bd5 100644 --- a/response_selectors/ranking_based_response_selector/README.md +++ b/response_selectors/ranking_based_response_selector/README.md @@ -1,3 +1,4 @@ +<<<<<<< HEAD # Ranking-based Response Selector ## Description @@ -16,3 +17,25 @@ Parameter `FILTER_TOXIC_OR_BADLISTED` defines whether it filers out toxic hypoth ## Dependencies - either candidate annotations by `SENTENCE_RANKER_ANNOTATION_NAME` or service `SENTENCE_RANKER_SERVICE_URL`. +======= +# Ranking_based_response_selector + +## Description +component_type: response selector +is_customizable: false + +## Input/Output + +**Input:** +text here + +**Output:** Ranking_based_response_selector service returns +- the selected skill’s name, +- the response text (which can be overwritten) +- the confidence level +``` + +``` + +## Dependencies +>>>>>>> readmes for dream multilingual dist From 1b76ab2267a20e281b6ff5835f479f5ec636b78b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Anast=C3=A1sis?= Date: Sun, 9 Jul 2023 03:59:35 +0400 Subject: [PATCH 06/28] main distribution readmes upd, TBC --- .../convers_evaluation_based_selector/README.md | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/response_selectors/convers_evaluation_based_selector/README.md b/response_selectors/convers_evaluation_based_selector/README.md index 026e19464d..2b188a040c 100644 --- a/response_selectors/convers_evaluation_based_selector/README.md +++ b/response_selectors/convers_evaluation_based_selector/README.md @@ -79,5 +79,9 @@ python response_selectors/convers_evaluation_based_selector/measure_quality.py \ Configuration settings specified in the .yaml files at [service_configs](service_configs) Required Python packages specified in [requirements.txt](requirements.txt). +<<<<<<< HEAD +For more details please refer to [this script](server.py) " +>>>>>>> main distribution readmes upd, TBC +======= For more details please refer to [this script](server.py) " >>>>>>> main distribution readmes upd, TBC From bb8ca2994bc361ca44974d2b938552b43eeca2a8 Mon Sep 17 00:00:00 2001 From: nstsj Date: Mon, 10 Jul 2023 12:56:43 +0400 Subject: [PATCH 07/28] readmes upd --- .../ranking_based_response_selector/README.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/response_selectors/ranking_based_response_selector/README.md b/response_selectors/ranking_based_response_selector/README.md index f028664bd5..fae0d1b7b3 100644 --- a/response_selectors/ranking_based_response_selector/README.md +++ b/response_selectors/ranking_based_response_selector/README.md @@ -21,8 +21,7 @@ Parameter `FILTER_TOXIC_OR_BADLISTED` defines whether it filers out toxic hypoth # Ranking_based_response_selector ## Description -component_type: response selector -is_customizable: false + ## Input/Output @@ -30,9 +29,10 @@ is_customizable: false text here **Output:** Ranking_based_response_selector service returns -- the selected skill’s name, +- the selected skill name, - the response text (which can be overwritten) - the confidence level + ``` ``` From 4f52edc3592691837c4d97f92ea48076a82d5ad2 Mon Sep 17 00:00:00 2001 From: nstsj Date: Mon, 10 Jul 2023 13:42:15 +0400 Subject: [PATCH 08/28] fixed readmes, those old ones with wrong format --- annotators/combined_classification/README.md | 1 - .../README.md | 36 ------------------- 2 files changed, 37 deletions(-) diff --git a/annotators/combined_classification/README.md b/annotators/combined_classification/README.md index f582b8a095..2b76ce1368 100644 --- a/annotators/combined_classification/README.md +++ b/annotators/combined_classification/README.md @@ -30,4 +30,3 @@ text here if i/o specified ## Dependencies -For more details please refer to [this script](server.py) \ No newline at end of file diff --git a/annotators/combined_classification_lightweight/README.md b/annotators/combined_classification_lightweight/README.md index c1e485b7e1..08aea83673 100644 --- a/annotators/combined_classification_lightweight/README.md +++ b/annotators/combined_classification_lightweight/README.md @@ -25,45 +25,9 @@ The models (multitask and comparative single task) were trained with initial lea This model(with a huawei-noah/TinyBERT_General_4L_312D backbone) on a CPU-only inference takes 42% less time than combined_classification, while using only ~1.5 Gb of the CPU instead of the 2909 Mb for combined_classification. The average accuracy and average F1 at the same time are for this model only ~1.5% lower than for the combined_classification, and this dropdown is consistent for all tasks. -# Component Name - -## Description -component_type: multitask model, NN-based -is_customizable: true - -This model is based on the transformer-agnostic multitask neural architecture. It can solve several tasks similtaneously, almost as good as single-task models. - -The models were trained on the following datasets: - -**Factoid classification** : For the Factoid task, we used the same Yahoo ConversVsInfo dataset that was used to train the Dream socialbot in Alexa Prize . Note that the valid set in this task was equal to the test set. - -**Midas classification** : For the Midas task, we used the same Midas classification dataset that was used to train the Dream socialbot in Alexa Prize . Note that the valid set in this task was equal to the test set. - -**Emotion classification** :For the Emotion classification task, we used the emo\_go\_emotions dataset, with all the 28 classes compressed into the seven basic emotions as in the original paper. Note that these 7 emotions are not exactly the same as the 7 emotions in the original Dream socialbot in Alexa Prize: 1 emotion differs (love VS disgust), so the scores are incomparable with the original model. Note that this task is multiclass. - -**Topic classification**: For the Topic classification task, we used the dataset made by Dilyara Zharikova. The dataset was further filtered and improved for the final model version, to make the model suitable for DREAM. Note that the original topics model doesn’t account for that dataset changes(which were also about class number) and thus its scores are not compatible with the scores we have. - -**Sentiment classification** : For the Sentiment classification task, we used the Dynabench dataset (r1 + r2). - -**Toxic classification** : For the toxic classification task, we used the dataset from kaggle the 7 toxic classes that pose an interest to us. Note that this task is multilabel. - -The model also contains 3 replacement models for Amazon services. - -The models (multitask and comparative single task) were trained with initial learning rate 2e-5(with validation patience 2 it could be dropped 2 times), batch size 32,optimizer adamW(betas (0.9,0.99) and early stop on 3 epochs. The criteria on early stopping was average accuracy for all tasks for multitask models, or the single-task accuracy for singletask models. - -This model(with a distilbert-base-uncased backbone) takes only 2439 Mb for 9 tasks, whereas single-task models with the same backbone for every of these tasks take up almost the same memory(~2437 Mb for every of these 9 tasks). - -## I/O -text here if i/o specified - -## Dependencies - -Configuration settings specified in the .yaml files at [service_configs](service_configs) -Required Python packages specified in [requirements.txt](requirements.txt) and [requirements_load_test](requirements_load_test.txt) ## I/O ## Dependencies -For more details please refer to [this script](server.py) From 7f05210ced8a93d10b29fe1c1870ad7ab1956488 Mon Sep 17 00:00:00 2001 From: nstsj Date: Mon, 10 Jul 2023 17:23:08 +0400 Subject: [PATCH 09/28] some more readmes --- annotators/spacy_nounphrases/README.md | 8 ++++++++ annotators/spacy_nounphrases/README.txt | 1 - services/image_captioning/README.md | 6 ++++++ skills/dff_template_skill/README.md | 12 ++++++++---- 4 files changed, 22 insertions(+), 5 deletions(-) create mode 100644 annotators/spacy_nounphrases/README.md delete mode 100644 annotators/spacy_nounphrases/README.txt diff --git a/annotators/spacy_nounphrases/README.md b/annotators/spacy_nounphrases/README.md new file mode 100644 index 0000000000..6d836bf672 --- /dev/null +++ b/annotators/spacy_nounphrases/README.md @@ -0,0 +1,8 @@ +# Spacy nounphrases annotator + +## Description +This is Cobot nounphrase annotator. + +## I/O + +## Dependencies \ No newline at end of file diff --git a/annotators/spacy_nounphrases/README.txt b/annotators/spacy_nounphrases/README.txt deleted file mode 100644 index 10a665d7f5..0000000000 --- a/annotators/spacy_nounphrases/README.txt +++ /dev/null @@ -1 +0,0 @@ -This is Cobot nounphrase annotator. diff --git a/services/image_captioning/README.md b/services/image_captioning/README.md index 8f43b4ff95..064515d1c5 100644 --- a/services/image_captioning/README.md +++ b/services/image_captioning/README.md @@ -1,4 +1,6 @@ # Image Captioning Service + +## Description **Image Captioning** is an OFA-based service that is used to get a text representation of a received image and respond accordingly. OFA is a unified multimodal pretrained model that unifies modalities and tasks to a simple sequence-to-sequence learning framework ([Unifying Architectures, Tasks, and Modalities Through a Simple Sequence-to-Sequence Learning Framework](http://arxiv.org/abs/2202.03052)). It also uses fairseq - a sequence modeling toolkit for training custom models for text generation tasks ([FAIRSEQ: A Fast, Extensible Toolkit for Sequence Modeling](https://aclanthology.org/N19-4009.pdf)). One 256 X 256 picture is processed ~0.7 sec (on average). @@ -16,3 +18,7 @@ sudo AGENT_PORT=4242 docker-compose -f docker-compose.yml -f assistant_dists/dre ```sh ./test.sh ``` + +## I/O + +## Dependencies diff --git a/skills/dff_template_skill/README.md b/skills/dff_template_skill/README.md index 6543ed4ca5..09625a9f84 100644 --- a/skills/dff_template_skill/README.md +++ b/skills/dff_template_skill/README.md @@ -1,4 +1,6 @@ # DialogFlow Framework Template + +## Description Changes can only be made in the `dialogflows` directory. Template has dialog flows based on programy (`repeating`) and based on valila python (`greeting`). @@ -74,8 +76,8 @@ def DFF_TEMPLATE_formatter(dialog: Dict) -> List[Dict]: ``` -# Tests -## Test creating +## Tests +### Test creating The file `server.py` contains this code @@ -128,7 +130,7 @@ except Exception as exc: - service testing by `test.sh` execution -## Test extending +### Test extending If you use service based on random behavior you can send `random_seed` in your service. You can find corespond lines in `server.py` ```python ... # some code @@ -299,4 +301,6 @@ Then in skill\_selectors/rule\_based_selector/connector.py: simplified_dialog_flow.set_error_successor(State.USR_INSERT_SMALLTALK, State.SYS_ERR) simplified_dialog_flow.set_error_successor(State.SYS_INSERT_FACT, State.SYS_ERR) simplified_dialog_flow.set_error_successor(State.USR_INSERT_FACT, State.SYS_ERR) -``` \ No newline at end of file +``` + +## Dependencies From 99be99d9e2e1f0f8a873000a210de0f32daf125d Mon Sep 17 00:00:00 2001 From: nstsj Date: Wed, 26 Jul 2023 14:57:25 +0400 Subject: [PATCH 10/28] added readme templates for the components that hadn't have them yet. Also added info about I/O to the existing readmes. TBA --- annotators/COMeT/README.md | 14 ++++++++++++++ annotators/ConversationEvaluator/README.md | 9 +++++++++ .../DeepPavlovFactoidClassification/README.md | 9 +++++++++ annotators/NER/README.md | 0 annotators/NER_deeppavlov/README.md | 0 annotators/SentRewrite/README.md | 0 annotators/SentSeg/README.md | 0 annotators/custom_entity_linking/README.md | 5 +++-- annotators/dialog_breakdown/README.md | 0 annotators/entity_detection/README.md | 0 annotators/entity_detection_rus/README.md | 0 annotators/entity_linking_deepy/README.md | 0 annotators/entity_linking_rus/README.md | 0 annotators/fact_random/README.md | 0 annotators/hypothesis_scorer/README.md | 0 annotators/midas_classification/README.md | 0 annotators/midas_predictor/README.md | 0 annotators/speech_function_classifier/README.md | 0 annotators/speech_function_predictor/README.md | 0 annotators/spelling_preprocessing_dp/README.md | 0 annotators/topic_recommendation/README.md | 0 annotators/user_persona_extractor/README.md | 9 +++++++++ 22 files changed, 44 insertions(+), 2 deletions(-) create mode 100644 annotators/ConversationEvaluator/README.md create mode 100644 annotators/DeepPavlovFactoidClassification/README.md create mode 100644 annotators/NER/README.md create mode 100644 annotators/NER_deeppavlov/README.md create mode 100644 annotators/SentRewrite/README.md create mode 100644 annotators/SentSeg/README.md create mode 100644 annotators/dialog_breakdown/README.md create mode 100644 annotators/entity_detection/README.md create mode 100644 annotators/entity_detection_rus/README.md create mode 100644 annotators/entity_linking_deepy/README.md create mode 100644 annotators/entity_linking_rus/README.md create mode 100644 annotators/fact_random/README.md create mode 100644 annotators/hypothesis_scorer/README.md create mode 100644 annotators/midas_classification/README.md create mode 100644 annotators/midas_predictor/README.md create mode 100644 annotators/speech_function_classifier/README.md create mode 100644 annotators/speech_function_predictor/README.md create mode 100644 annotators/spelling_preprocessing_dp/README.md create mode 100644 annotators/topic_recommendation/README.md create mode 100644 annotators/user_persona_extractor/README.md diff --git a/annotators/COMeT/README.md b/annotators/COMeT/README.md index 978ae9a63c..a0fca9ff75 100644 --- a/annotators/COMeT/README.md +++ b/annotators/COMeT/README.md @@ -36,6 +36,20 @@ docker-compose -f docker-compose.yml -f local.yml exec comet-conceptnet bash tes | Average starting time | 4s | 3s | | Average request execution time | 0.4s | 0.2s | +## Input/Output + +**Input** +- hypotheses: possible assistant's replies (?) +- currentUtterance: latest reply from a user +- pastResponses: a list of user utterances + +**Output** +a list of probabilities about the utterance: +- isResponseComprehensible +- isResponseErroneous +- isResponseInteresting +- isResponseOnTopic +- responseEngagesUser ## Dependencies diff --git a/annotators/ConversationEvaluator/README.md b/annotators/ConversationEvaluator/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/annotators/ConversationEvaluator/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/annotators/DeepPavlovFactoidClassification/README.md b/annotators/DeepPavlovFactoidClassification/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/annotators/DeepPavlovFactoidClassification/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/annotators/NER/README.md b/annotators/NER/README.md new file mode 100644 index 0000000000..e69de29bb2 diff --git a/annotators/NER_deeppavlov/README.md b/annotators/NER_deeppavlov/README.md new file mode 100644 index 0000000000..e69de29bb2 diff --git a/annotators/SentRewrite/README.md b/annotators/SentRewrite/README.md new file mode 100644 index 0000000000..e69de29bb2 diff --git a/annotators/SentSeg/README.md b/annotators/SentSeg/README.md new file mode 100644 index 0000000000..e69de29bb2 diff --git a/annotators/custom_entity_linking/README.md b/annotators/custom_entity_linking/README.md index e6477d14e4..7a6d833929 100644 --- a/annotators/custom_entity_linking/README.md +++ b/annotators/custom_entity_linking/README.md @@ -20,10 +20,11 @@ Relation examples: ## I/O **Inpunt** - +user_id, entity substring, entity_tags **Output:** -the annotator returns: +the annotator returns the processed entity information about: + - entities - entity_id (ids for multiple entities) - entity_confidence score diff --git a/annotators/dialog_breakdown/README.md b/annotators/dialog_breakdown/README.md new file mode 100644 index 0000000000..e69de29bb2 diff --git a/annotators/entity_detection/README.md b/annotators/entity_detection/README.md new file mode 100644 index 0000000000..e69de29bb2 diff --git a/annotators/entity_detection_rus/README.md b/annotators/entity_detection_rus/README.md new file mode 100644 index 0000000000..e69de29bb2 diff --git a/annotators/entity_linking_deepy/README.md b/annotators/entity_linking_deepy/README.md new file mode 100644 index 0000000000..e69de29bb2 diff --git a/annotators/entity_linking_rus/README.md b/annotators/entity_linking_rus/README.md new file mode 100644 index 0000000000..e69de29bb2 diff --git a/annotators/fact_random/README.md b/annotators/fact_random/README.md new file mode 100644 index 0000000000..e69de29bb2 diff --git a/annotators/hypothesis_scorer/README.md b/annotators/hypothesis_scorer/README.md new file mode 100644 index 0000000000..e69de29bb2 diff --git a/annotators/midas_classification/README.md b/annotators/midas_classification/README.md new file mode 100644 index 0000000000..e69de29bb2 diff --git a/annotators/midas_predictor/README.md b/annotators/midas_predictor/README.md new file mode 100644 index 0000000000..e69de29bb2 diff --git a/annotators/speech_function_classifier/README.md b/annotators/speech_function_classifier/README.md new file mode 100644 index 0000000000..e69de29bb2 diff --git a/annotators/speech_function_predictor/README.md b/annotators/speech_function_predictor/README.md new file mode 100644 index 0000000000..e69de29bb2 diff --git a/annotators/spelling_preprocessing_dp/README.md b/annotators/spelling_preprocessing_dp/README.md new file mode 100644 index 0000000000..e69de29bb2 diff --git a/annotators/topic_recommendation/README.md b/annotators/topic_recommendation/README.md new file mode 100644 index 0000000000..e69de29bb2 diff --git a/annotators/user_persona_extractor/README.md b/annotators/user_persona_extractor/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/annotators/user_persona_extractor/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file From 3bcb9bd49dfcfe8b29d6a3cfc37c85404dd36a32 Mon Sep 17 00:00:00 2001 From: nstsj Date: Wed, 25 Oct 2023 22:14:53 +0200 Subject: [PATCH 11/28] readme upd for skills -- templates added --- annotators/NER/README.md | 9 +++++++++ annotators/NER_deeppavlov/README.md | 9 +++++++++ annotators/SentRewrite/README.md | 9 +++++++++ annotators/SentSeg/README.md | 9 +++++++++ annotators/asr/README.md | 8 ++++++++ annotators/dialog_breakdown/README.md | 9 +++++++++ annotators/entity_detection/README.md | 9 +++++++++ annotators/entity_detection_rus/README.md | 9 +++++++++ annotators/entity_linking_deepy/README.md | 9 +++++++++ annotators/entity_linking_rus/README.md | 9 +++++++++ annotators/fact_random/README.md | 9 +++++++++ annotators/hypothesis_scorer/README.md | 9 +++++++++ annotators/midas_classification/README.md | 9 +++++++++ annotators/midas_predictor/README.md | 9 +++++++++ annotators/speech_function_classifier/README.md | 9 +++++++++ annotators/speech_function_predictor/README.md | 9 +++++++++ annotators/spelling_preprocessing_dp/README.md | 9 +++++++++ annotators/topic_recommendation/README.md | 9 +++++++++ .../description_based_skill_selector/README.md | 2 +- skill_selectors/post_annotator_selector/README.md | 9 +++++++++ skills/alexa_handler/README.md | 9 +++++++++ skills/christmas_new_year_skill/README.md | 9 +++++++++ skills/convert_reddit/README.md | 9 +++++++++ skills/convert_reddit_with_personality/README.md | 9 +++++++++ skills/dummy_skill_dialog/README.md | 9 +++++++++ skills/eliza/README.md | 9 +++++++++ skills/factoid_qa/README.md | 9 +++++++++ skills/faq_skill_deepy/README.md | 9 +++++++++ skills/game_cooperative_skill/README.md | 9 +++++++++ .../harvesters_maintenance_gobot_skill_deepy/README.md | 9 +++++++++ skills/harvesters_maintenance_skill_deepy/README.md | 9 +++++++++ skills/knowledge_grounding_skill/README.md | 9 +++++++++ skills/meta_script_skill/README.md | 9 +++++++++ skills/misheard_asr/README.md | 9 +++++++++ skills/news_api_skill/README.md | 9 +++++++++ skills/oscar_skill/README.md | 9 +++++++++ skills/personal_info_skill/README.md | 9 +++++++++ skills/program_y_deepy/README.md | 9 +++++++++ skills/retrieval_chitchat/README.md | 9 +++++++++ skills/small_talk_skill/README.md | 9 +++++++++ skills/superbowl_skill/README.md | 9 +++++++++ skills/topicalchat_convert_retrieval/README.md | 9 +++++++++ skills/topicalchat_tfidf_retrieval/README.md | 9 +++++++++ skills/valentines_day_skill/README.md | 9 +++++++++ 44 files changed, 387 insertions(+), 1 deletion(-) create mode 100644 skill_selectors/post_annotator_selector/README.md create mode 100644 skills/alexa_handler/README.md create mode 100644 skills/christmas_new_year_skill/README.md create mode 100644 skills/convert_reddit/README.md create mode 100644 skills/convert_reddit_with_personality/README.md create mode 100644 skills/dummy_skill_dialog/README.md create mode 100644 skills/eliza/README.md create mode 100644 skills/factoid_qa/README.md create mode 100644 skills/faq_skill_deepy/README.md create mode 100644 skills/game_cooperative_skill/README.md create mode 100644 skills/harvesters_maintenance_gobot_skill_deepy/README.md create mode 100644 skills/harvesters_maintenance_skill_deepy/README.md create mode 100644 skills/knowledge_grounding_skill/README.md create mode 100644 skills/meta_script_skill/README.md create mode 100644 skills/misheard_asr/README.md create mode 100644 skills/news_api_skill/README.md create mode 100644 skills/oscar_skill/README.md create mode 100644 skills/personal_info_skill/README.md create mode 100644 skills/program_y_deepy/README.md create mode 100644 skills/retrieval_chitchat/README.md create mode 100644 skills/small_talk_skill/README.md create mode 100644 skills/superbowl_skill/README.md create mode 100644 skills/topicalchat_convert_retrieval/README.md create mode 100644 skills/topicalchat_tfidf_retrieval/README.md create mode 100644 skills/valentines_day_skill/README.md diff --git a/annotators/NER/README.md b/annotators/NER/README.md index e69de29bb2..9a5a832efd 100644 --- a/annotators/NER/README.md +++ b/annotators/NER/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/annotators/NER_deeppavlov/README.md b/annotators/NER_deeppavlov/README.md index e69de29bb2..9a5a832efd 100644 --- a/annotators/NER_deeppavlov/README.md +++ b/annotators/NER_deeppavlov/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/annotators/SentRewrite/README.md b/annotators/SentRewrite/README.md index e69de29bb2..9a5a832efd 100644 --- a/annotators/SentRewrite/README.md +++ b/annotators/SentRewrite/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/annotators/SentSeg/README.md b/annotators/SentSeg/README.md index e69de29bb2..9a5a832efd 100644 --- a/annotators/SentSeg/README.md +++ b/annotators/SentSeg/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/annotators/asr/README.md b/annotators/asr/README.md index 42f3a12684..32ff8faeab 100644 --- a/annotators/asr/README.md +++ b/annotators/asr/README.md @@ -4,7 +4,15 @@ ASR component allows users to provide speech input via its `http://_service_name_:4343/asr?user_id=` endpoint. To do so, attach the recorded voice as a `.wav` file, 16KHz. +This component calculates overall ASR confidence for a given utterance and grades it as either *very low*, *low*, *medium*, *high* or *undefined* (for Amazon markup). + ## I/O +**Input:** +user utterance: recorded voice as a `.wav` file + +**Output** +asr_confidence: a probability of a user speech recognition ## Dependencies + diff --git a/annotators/dialog_breakdown/README.md b/annotators/dialog_breakdown/README.md index e69de29bb2..9a5a832efd 100644 --- a/annotators/dialog_breakdown/README.md +++ b/annotators/dialog_breakdown/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/annotators/entity_detection/README.md b/annotators/entity_detection/README.md index e69de29bb2..9a5a832efd 100644 --- a/annotators/entity_detection/README.md +++ b/annotators/entity_detection/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/annotators/entity_detection_rus/README.md b/annotators/entity_detection_rus/README.md index e69de29bb2..9a5a832efd 100644 --- a/annotators/entity_detection_rus/README.md +++ b/annotators/entity_detection_rus/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/annotators/entity_linking_deepy/README.md b/annotators/entity_linking_deepy/README.md index e69de29bb2..9a5a832efd 100644 --- a/annotators/entity_linking_deepy/README.md +++ b/annotators/entity_linking_deepy/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/annotators/entity_linking_rus/README.md b/annotators/entity_linking_rus/README.md index e69de29bb2..9a5a832efd 100644 --- a/annotators/entity_linking_rus/README.md +++ b/annotators/entity_linking_rus/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/annotators/fact_random/README.md b/annotators/fact_random/README.md index e69de29bb2..9a5a832efd 100644 --- a/annotators/fact_random/README.md +++ b/annotators/fact_random/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/annotators/hypothesis_scorer/README.md b/annotators/hypothesis_scorer/README.md index e69de29bb2..9a5a832efd 100644 --- a/annotators/hypothesis_scorer/README.md +++ b/annotators/hypothesis_scorer/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/annotators/midas_classification/README.md b/annotators/midas_classification/README.md index e69de29bb2..9a5a832efd 100644 --- a/annotators/midas_classification/README.md +++ b/annotators/midas_classification/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/annotators/midas_predictor/README.md b/annotators/midas_predictor/README.md index e69de29bb2..9a5a832efd 100644 --- a/annotators/midas_predictor/README.md +++ b/annotators/midas_predictor/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/annotators/speech_function_classifier/README.md b/annotators/speech_function_classifier/README.md index e69de29bb2..9a5a832efd 100644 --- a/annotators/speech_function_classifier/README.md +++ b/annotators/speech_function_classifier/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/annotators/speech_function_predictor/README.md b/annotators/speech_function_predictor/README.md index e69de29bb2..9a5a832efd 100644 --- a/annotators/speech_function_predictor/README.md +++ b/annotators/speech_function_predictor/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/annotators/spelling_preprocessing_dp/README.md b/annotators/spelling_preprocessing_dp/README.md index e69de29bb2..9a5a832efd 100644 --- a/annotators/spelling_preprocessing_dp/README.md +++ b/annotators/spelling_preprocessing_dp/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/annotators/topic_recommendation/README.md b/annotators/topic_recommendation/README.md index e69de29bb2..9a5a832efd 100644 --- a/annotators/topic_recommendation/README.md +++ b/annotators/topic_recommendation/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/skill_selectors/description_based_skill_selector/README.md b/skill_selectors/description_based_skill_selector/README.md index 03e8fdb173..2356fba598 100644 --- a/skill_selectors/description_based_skill_selector/README.md +++ b/skill_selectors/description_based_skill_selector/README.md @@ -25,4 +25,4 @@ ALWAYS_TURN_ON_ALL_SKILLS: whether to always turn on all skills. Avoid using thi ## Dependencies -- Prompt Selector annotations. If not provided, Skill Selector turns on all prompt-based skills from `pipeline_conf.json` (skills section). \ No newline at end of file +- Prompt Selector annotations. If not provided, Skill Selector turns on all prompt-based skills from `pipeline_conf.json` (skills section). diff --git a/skill_selectors/post_annotator_selector/README.md b/skill_selectors/post_annotator_selector/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/skill_selectors/post_annotator_selector/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/skills/alexa_handler/README.md b/skills/alexa_handler/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/skills/alexa_handler/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/skills/christmas_new_year_skill/README.md b/skills/christmas_new_year_skill/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/skills/christmas_new_year_skill/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/skills/convert_reddit/README.md b/skills/convert_reddit/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/skills/convert_reddit/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/skills/convert_reddit_with_personality/README.md b/skills/convert_reddit_with_personality/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/skills/convert_reddit_with_personality/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/skills/dummy_skill_dialog/README.md b/skills/dummy_skill_dialog/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/skills/dummy_skill_dialog/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/skills/eliza/README.md b/skills/eliza/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/skills/eliza/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/skills/factoid_qa/README.md b/skills/factoid_qa/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/skills/factoid_qa/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/skills/faq_skill_deepy/README.md b/skills/faq_skill_deepy/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/skills/faq_skill_deepy/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/skills/game_cooperative_skill/README.md b/skills/game_cooperative_skill/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/skills/game_cooperative_skill/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/skills/harvesters_maintenance_gobot_skill_deepy/README.md b/skills/harvesters_maintenance_gobot_skill_deepy/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/skills/harvesters_maintenance_gobot_skill_deepy/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/skills/harvesters_maintenance_skill_deepy/README.md b/skills/harvesters_maintenance_skill_deepy/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/skills/harvesters_maintenance_skill_deepy/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/skills/knowledge_grounding_skill/README.md b/skills/knowledge_grounding_skill/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/skills/knowledge_grounding_skill/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/skills/meta_script_skill/README.md b/skills/meta_script_skill/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/skills/meta_script_skill/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/skills/misheard_asr/README.md b/skills/misheard_asr/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/skills/misheard_asr/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/skills/news_api_skill/README.md b/skills/news_api_skill/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/skills/news_api_skill/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/skills/oscar_skill/README.md b/skills/oscar_skill/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/skills/oscar_skill/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/skills/personal_info_skill/README.md b/skills/personal_info_skill/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/skills/personal_info_skill/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/skills/program_y_deepy/README.md b/skills/program_y_deepy/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/skills/program_y_deepy/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/skills/retrieval_chitchat/README.md b/skills/retrieval_chitchat/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/skills/retrieval_chitchat/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/skills/small_talk_skill/README.md b/skills/small_talk_skill/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/skills/small_talk_skill/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/skills/superbowl_skill/README.md b/skills/superbowl_skill/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/skills/superbowl_skill/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/skills/topicalchat_convert_retrieval/README.md b/skills/topicalchat_convert_retrieval/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/skills/topicalchat_convert_retrieval/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/skills/topicalchat_tfidf_retrieval/README.md b/skills/topicalchat_tfidf_retrieval/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/skills/topicalchat_tfidf_retrieval/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file diff --git a/skills/valentines_day_skill/README.md b/skills/valentines_day_skill/README.md new file mode 100644 index 0000000000..9a5a832efd --- /dev/null +++ b/skills/valentines_day_skill/README.md @@ -0,0 +1,9 @@ +# Title +## Description + +## Input/Output + +**Input** +**Output** + +## Dependencies \ No newline at end of file From b771590edbe95f6c4cb71e3e35d1794bf2b3276a Mon Sep 17 00:00:00 2001 From: nstsj Date: Wed, 25 Oct 2023 22:15:14 +0200 Subject: [PATCH 12/28] readme upd for skills -- templates added --- annotators/property_extraction/README.md | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/annotators/property_extraction/README.md b/annotators/property_extraction/README.md index 483914948e..a2f651b55e 100644 --- a/annotators/property_extraction/README.md +++ b/annotators/property_extraction/README.md @@ -27,3 +27,7 @@ requests.post("http://0.0.0.0:8136/respond", json = {"utterances": utterances}). {"triplets": [{"subject": "user", "property": "marital status", "object": "husband"}, {"subject": "user", "relation": "like activity", "object": "travel"}]} ] ``` + +**Output** + +## Dependencies \ No newline at end of file From 1f3be1f31938dd67de913173a956d0285b043e1a Mon Sep 17 00:00:00 2001 From: nstsj Date: Thu, 27 Jul 2023 19:34:23 +0400 Subject: [PATCH 13/28] fixed readmes --- annotators/BadlistedWordsDetector/README.md | 2 +- annotators/BadlistedWordsDetector_ru/README.md | 2 +- annotators/COMeT/README.md | 1 + annotators/combined_classification/README.md | 1 - annotators/combined_classification_lightweight/README.md | 9 +-------- 5 files changed, 4 insertions(+), 11 deletions(-) diff --git a/annotators/BadlistedWordsDetector/README.md b/annotators/BadlistedWordsDetector/README.md index 87f23e369e..849dd87896 100644 --- a/annotators/BadlistedWordsDetector/README.md +++ b/annotators/BadlistedWordsDetector/README.md @@ -8,4 +8,4 @@ input: "sentences": ["fucking hell", "he mishit the shot", "you asshole"], output: words and their tags [{"bad_words": True}, {"bad_words": False}, {"bad_words": True}] -## Dependencies +## Dependencies \ No newline at end of file diff --git a/annotators/BadlistedWordsDetector_ru/README.md b/annotators/BadlistedWordsDetector_ru/README.md index b59292528f..a111ea627c 100644 --- a/annotators/BadlistedWordsDetector_ru/README.md +++ b/annotators/BadlistedWordsDetector_ru/README.md @@ -1,4 +1,4 @@ -# BadlistedWordsDetector +# BadlistedWordsDetector for Russian ## Description diff --git a/annotators/COMeT/README.md b/annotators/COMeT/README.md index a0fca9ff75..baed543fff 100644 --- a/annotators/COMeT/README.md +++ b/annotators/COMeT/README.md @@ -5,6 +5,7 @@ COMeT is a Commonsense Transformers for Automatic Knowledge Graph Construction service based on [comet-commonsense](https://github.com/atcbosselut/comet-commonsense) framework written in Python 3. + ### Quickstart from docker for COMeT with Atomic graph ```bash diff --git a/annotators/combined_classification/README.md b/annotators/combined_classification/README.md index 2b76ce1368..10bc233021 100644 --- a/annotators/combined_classification/README.md +++ b/annotators/combined_classification/README.md @@ -29,4 +29,3 @@ This model(with a distilbert-base-uncased backbone) takes only 2439 Mb for 9 tas text here if i/o specified ## Dependencies - diff --git a/annotators/combined_classification_lightweight/README.md b/annotators/combined_classification_lightweight/README.md index 08aea83673..ef727e57f9 100644 --- a/annotators/combined_classification_lightweight/README.md +++ b/annotators/combined_classification_lightweight/README.md @@ -23,11 +23,4 @@ The model also contains 3 replacement models for Amazon services. The models (multitask and comparative single task) were trained with initial learning rate 2e-5(with validation patience 2 it could be dropped 2 times), batch size 32,optimizer adamW(betas (0.9,0.99) and early stop on 3 epochs. The criteria on early stopping was average accuracy for all tasks for multitask models, or the single-task accuracy for singletask models. -This model(with a huawei-noah/TinyBERT_General_4L_312D backbone) on a CPU-only inference takes 42% less time than combined_classification, while using only ~1.5 Gb of the CPU instead of the 2909 Mb for combined_classification. The average accuracy and average F1 at the same time are for this model only ~1.5% lower than for the combined_classification, and this dropdown is consistent for all tasks. - - -## I/O - - -## Dependencies - +This model(with a huawei-noah/TinyBERT_General_4L_312D backbone) on a CPU-only inference takes 42% less time than combined_classification, while using only ~1.5 Gb of the CPU instead of the 2909 Mb for combined_classification. The average accuracy and average F1 at the same time are for this model only ~1.5% lower than for the combined_classification, and this dropdown is consistent for all tasks. \ No newline at end of file From 6fc1affbd57714b79be19a209635dafd45859d6d Mon Sep 17 00:00:00 2001 From: nstsj Date: Sat, 29 Jul 2023 15:38:05 +0400 Subject: [PATCH 14/28] upd readmes in skill selectors and response selectors --- .../README.md | 57 +++---------------- .../post_annotator_selector/README.md | 17 +++++- skill_selectors/rule_based_selector/README.md | 15 +++-- 3 files changed, 33 insertions(+), 56 deletions(-) diff --git a/response_selectors/convers_evaluation_based_selector/README.md b/response_selectors/convers_evaluation_based_selector/README.md index 2b188a040c..0e74ac3166 100644 --- a/response_selectors/convers_evaluation_based_selector/README.md +++ b/response_selectors/convers_evaluation_based_selector/README.md @@ -3,9 +3,11 @@ ## Description Response Selector is a component selecting final response among the given hypotheses by different skills. + The Tag- and Evaluation--based Response Selector utilizes a complicated approach which aims to -prioritize scripted skills while having an opportunity to provide a system-initiative via so called linking questions -that bring conversation to the scripts. A final hypotheses could be a combination of a hypotheses and linking question. +prioritize scripted skills while having an opportunity to provide a system-initiative via so-called linking questions that bring conversation to the scripts. +A final hypothesis could be a combination of a hypothesis and linking question. + The approach is most suitable for distributions where the most of the responses are implied to be by scripts. ### Parameters @@ -36,52 +38,9 @@ LANGUAGE: language to consider FALLBACK_FILE: a file name with fallbacks from `dream/common/fallbacks/` ``` -## Dependencies - -<<<<<<< HEAD - -======= -For this context you may type 0,3 hypots: -``` -Type best hypot num(s), separated by comma: 0,3 -``` - -When dialog ends it saves it into labeled_data folder. - -### Measuring quality - -It takes `--data_dir` as an input (the same as save_dir from data labelling script). -Outputs overall accuracy. - -Example of usage: - -``` -python response_selectors/convers_evaluation_based_selector/measure_quality.py \ - --data_dir response_selectors/convers_evaluation_based_selector/labeled_data/ - -Overall accuracy: 0.5185185185185185 -``` - -### How to run conversation evaluator locally - -`docker-compose -f docker-compose.yml -f dev.yml -f cpu.yml -f one_worker.yml up toxic_classification badlisted_words convers_evaluation_selector` +## Input/Output +**Input:** a list of hypotheses with their scores and metadata +**Output:** a list of selected hypotheses with their scores and metadata -Then use `--url`. - -Example of usage with url: - -``` -python response_selectors/convers_evaluation_based_selector/measure_quality.py \ - --data_dir response_selectors/convers_evaluation_based_selector/labeled_data/ \ - --url http://0.0.0.0:8009/respond -``` ## Dependencies - -Configuration settings specified in the .yaml files at [service_configs](service_configs) -Required Python packages specified in [requirements.txt](requirements.txt). -<<<<<<< HEAD -For more details please refer to [this script](server.py) " ->>>>>>> main distribution readmes upd, TBC -======= -For more details please refer to [this script](server.py) " ->>>>>>> main distribution readmes upd, TBC +none diff --git a/skill_selectors/post_annotator_selector/README.md b/skill_selectors/post_annotator_selector/README.md index 9a5a832efd..0eb21a6c93 100644 --- a/skill_selectors/post_annotator_selector/README.md +++ b/skill_selectors/post_annotator_selector/README.md @@ -1,9 +1,22 @@ -# Title +# Post-Annotator Skill Selector + + ## Description +The Skill Selector service provides a list of selected skills to generate a response for a dialogue. It is a part of the main Dream distribution, built on the DeepPavlov Agent framework. + +The Skill Selector forms a list of the most relevant skills based on the dialogue context. +It considers fallback and open-domain skills by default. Closed-domain skills are selected if specific triggers _(topics, entities, intents, or regular expressions)_ are detected. + +To avoid resource overload, the number of selected skills is controlled. + ## Input/Output **Input** +Annotated user input and dialogue context + **Output** +A list of selected skills -## Dependencies \ No newline at end of file +## Dependencies +none \ No newline at end of file diff --git a/skill_selectors/rule_based_selector/README.md b/skill_selectors/rule_based_selector/README.md index be70526080..c47a026975 100644 --- a/skill_selectors/rule_based_selector/README.md +++ b/skill_selectors/rule_based_selector/README.md @@ -3,12 +3,16 @@ ## Description Skill Selector is a component selecting a subset of skills to generate hypotheses. -The Rule-based Skill Selector utilizes hand-written rules based on user's utterance annotations -(e.g., entities, dialog acts, intents, topics, toxicity, etc.). -Best fit for Dream Scripted, Dream Alexa distributions and other ones containing a lot of scripted skills. -Not suitable for prompt-based distributions. +The Rule-based Skill Selector utilizes hand-written rules based on user's utterance annotations _(entities, dialog acts, intents, topics, toxicity, etc.)_. +Best fitted for `Dream Scripted`, `Dream Alexa` distributions and other ones containing a lot of scripted skills. +**Not suitable** for prompt-based distributions. -### Parameters +## Input/Output + +**Input:** annotated user input and dialogue context +**Output:** a list of selected skills + +## Parameters ``` HIGH_PRIORITY_INTENTS: whether to turn on only Intent Responder for high-priority intents @@ -17,3 +21,4 @@ ALWAYS_TURN_ON_ALL_SKILLS: whether to always turn on all skills. Avoid using thi ``` ## Dependencies +none From e4a467a73b2fdc6e00134f6d1a410c4bd8600037 Mon Sep 17 00:00:00 2001 From: nstsj Date: Fri, 4 Aug 2023 13:33:16 +0400 Subject: [PATCH 15/28] upd in dependencies and I/O --- annotators/BadlistedWordsDetector/README.md | 5 ++-- .../BadlistedWordsDetector_ru/README.md | 5 ++-- annotators/COMeT/README.md | 6 +++-- annotators/ConversationEvaluator/README.md | 17 ++++++++++-- .../DeepPavlovEmotionClassification/README.md | 16 +++++++++++- annotators/asr/README.md | 1 + annotators/combined_classification/README.md | 7 +++-- .../README.md | 13 ++++++++-- annotators/custom_entity_linking/README.md | 26 +++++++++---------- 9 files changed, 69 insertions(+), 27 deletions(-) diff --git a/annotators/BadlistedWordsDetector/README.md b/annotators/BadlistedWordsDetector/README.md index 849dd87896..e5991f86ed 100644 --- a/annotators/BadlistedWordsDetector/README.md +++ b/annotators/BadlistedWordsDetector/README.md @@ -4,8 +4,7 @@ Spacy-based user utterance annotator that detects words and phrases from the badlist ## I/O -input: "sentences": ["fucking hell", "he mishit the shot", "you asshole"], -output: words and their tags +**Input:** a list of user's uttetances [{"bad_words": True}, {"bad_words": False}, {"bad_words": True}] -## Dependencies \ No newline at end of file +## Dependencies diff --git a/annotators/BadlistedWordsDetector_ru/README.md b/annotators/BadlistedWordsDetector_ru/README.md index a111ea627c..d2e32a5b0d 100644 --- a/annotators/BadlistedWordsDetector_ru/README.md +++ b/annotators/BadlistedWordsDetector_ru/README.md @@ -5,7 +5,8 @@ Spacy-based user utterance annotator that detects words and phrases from the badlist. This version of the annotator works for the Russian Language. ## I/O -input: user input as a str, lang = ru -output: json dict +**Input:** a list of user's uttetances +**Output:** words and their tags (`{"bad_words": False}` or `{"bad_words": True}`) ## Dependencies +none \ No newline at end of file diff --git a/annotators/COMeT/README.md b/annotators/COMeT/README.md index baed543fff..41034846c3 100644 --- a/annotators/COMeT/README.md +++ b/annotators/COMeT/README.md @@ -40,9 +40,9 @@ docker-compose -f docker-compose.yml -f local.yml exec comet-conceptnet bash tes ## Input/Output **Input** -- hypotheses: possible assistant's replies (?) +- hypotheses: possible assistant's replies - currentUtterance: latest reply from a user -- pastResponses: a list of user utterances +- pastResponses: a list of user's utterances **Output** a list of probabilities about the utterance: @@ -54,3 +54,5 @@ a list of probabilities about the utterance: ## Dependencies + +none \ No newline at end of file diff --git a/annotators/ConversationEvaluator/README.md b/annotators/ConversationEvaluator/README.md index 9a5a832efd..90767c4a2b 100644 --- a/annotators/ConversationEvaluator/README.md +++ b/annotators/ConversationEvaluator/README.md @@ -1,9 +1,22 @@ -# Title +# Conversation Evaluator + ## Description +This annotator is trained on the Alexa Prize data from the previous competitions and predicts whether the candidate response is interesting, comprehensible, on-topic, engaging, or erroneous. ## Input/Output **Input** +- possible assistant's replies +- user's past responses **Output** +tags +- `isResponseComprehensible` +- `isResponseErroneous` +- `isResponseInteresting` +- `isResponseOnTopic` +- `responseEngagesUser` + +with their probabilities -## Dependencies \ No newline at end of file +## Dependencies +none \ No newline at end of file diff --git a/annotators/DeepPavlovEmotionClassification/README.md b/annotators/DeepPavlovEmotionClassification/README.md index ee708f9801..1957519484 100644 --- a/annotators/DeepPavlovEmotionClassification/README.md +++ b/annotators/DeepPavlovEmotionClassification/README.md @@ -1 +1,15 @@ -BERT Base model for emotion classification which learned at the custom dataset(described more precisely in our article) +# DeepPavlov Emotion Classification Annotator + +## Description + +BERT Base model for emotion classification which learned at the custom dataset (described more precisely in (our article)[link here]) + +## I/O + +**Inpunt** + +**Output:** + + +## Dependencies +none \ No newline at end of file diff --git a/annotators/asr/README.md b/annotators/asr/README.md index 32ff8faeab..4ce40764d3 100644 --- a/annotators/asr/README.md +++ b/annotators/asr/README.md @@ -15,4 +15,5 @@ asr_confidence: a probability of a user speech recognition ## Dependencies +none diff --git a/annotators/combined_classification/README.md b/annotators/combined_classification/README.md index 10bc233021..e72d52ad15 100644 --- a/annotators/combined_classification/README.md +++ b/annotators/combined_classification/README.md @@ -21,11 +21,14 @@ The models were trained on the following datasets: The model also contains 3 replacement models for Amazon services. -The models (multitask and comparative single task) were trained with initial learning rate 2e-5(with validation patience 2 it could be dropped 2 times), batch size 32,optimizer adamW(betas (0.9,0.99) and early stop on 3 epochs. The criteria on early stopping was average accuracy for all tasks for multitask models, or the single-task accuracy for singletask models. +The models (multitask and comparative single task) were trained with initial learning rate 2e-5(with validation patience 2 it could be dropped 2 times), batch size 32,optimizer adamW(betas (0.9,0.99)) and early stop on 3 epochs. The criteria on early stopping was average accuracy for all tasks for multitask models, or the single-task accuracy for singletask models. This model(with a distilbert-base-uncased backbone) takes only 2439 Mb for 9 tasks, whereas single-task models with the same backbone for every of these tasks take up almost the same memory(~2437 Mb for every of these 9 tasks). ## I/O -text here if i/o specified +**Input:** immediate user utterances (+ optional history of previous utterances) +**Output:** tags for each utterance (based on toxic/topic/emotion/sentiment/factoid/midas classification) ## Dependencies +none + diff --git a/annotators/combined_classification_lightweight/README.md b/annotators/combined_classification_lightweight/README.md index ef727e57f9..4ee4d2241a 100644 --- a/annotators/combined_classification_lightweight/README.md +++ b/annotators/combined_classification_lightweight/README.md @@ -21,6 +21,15 @@ The models were trained on the following datasets: The model also contains 3 replacement models for Amazon services. -The models (multitask and comparative single task) were trained with initial learning rate 2e-5(with validation patience 2 it could be dropped 2 times), batch size 32,optimizer adamW(betas (0.9,0.99) and early stop on 3 epochs. The criteria on early stopping was average accuracy for all tasks for multitask models, or the single-task accuracy for singletask models. +The models (multitask and comparative single task) were trained with initial learning rate 2e-5(with validation patience 2 it could be dropped 2 times), batch size 32,optimizer adamW(betas (0.9,0.99)) and early stop on 3 epochs. The criteria on early stopping was average accuracy for all tasks for multitask models, or the single-task accuracy for singletask models. + +This model(with a huawei-noah/TinyBERT_General_4L_312D backbone) on a CPU-only inference takes 42% less time than combined_classification, while using only ~1.5 Gb of the CPU instead of the 2909 Mb for combined_classification. The average accuracy and average F1 at the same time are for this model only ~1.5% lower than for the combined_classification, and this dropdown is consistent for all tasks. + + +## I/O +**Input:** immediate user utterances (+ optional history of previous utterances) +**Output:** tags for each utterance (based on toxic/topic/emotion/sentiment/factoid/midas classification) + +## Dependencies +none -This model(with a huawei-noah/TinyBERT_General_4L_312D backbone) on a CPU-only inference takes 42% less time than combined_classification, while using only ~1.5 Gb of the CPU instead of the 2909 Mb for combined_classification. The average accuracy and average F1 at the same time are for this model only ~1.5% lower than for the combined_classification, and this dropdown is consistent for all tasks. \ No newline at end of file diff --git a/annotators/custom_entity_linking/README.md b/annotators/custom_entity_linking/README.md index 7a6d833929..759da898f3 100644 --- a/annotators/custom_entity_linking/README.md +++ b/annotators/custom_entity_linking/README.md @@ -4,17 +4,17 @@ This component is an Annotator that sematically links entities detected in user utterances. Entites then bound via relations. Relation examples: -- favorite animal -- like animal -- favorite book -- like read -- favorite movie -- favorite food -- like food -- favorite drink -- like drink -- favorite sport -- like sports +- `favorite animal` +- `like animal` +- `favorite book` +- `like read` +- `favorite movie` +- `favorite food` +- `like food` +- `favorite drink` +- `like drink` +- `favorite sport` +- `like sports` ## I/O @@ -23,11 +23,11 @@ Relation examples: user_id, entity substring, entity_tags **Output:** -the annotator returns the processed entity information about: - +processed information about: - entities - entity_id (ids for multiple entities) - entity_confidence score - entity_id_tags ## Dependencies + annotators: `NER`, `entity_detection` and `spacy_nounphrases` From 7d7e41a0aa3c0ff88ecb8c698b3c95c7ffc35eeb Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Anast=C3=A1sis?= Date: Sun, 9 Jul 2023 03:44:48 +0400 Subject: [PATCH 16/28] readmes for dream multilingual dist --- .../ranking_based_response_selector/README.md | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/response_selectors/ranking_based_response_selector/README.md b/response_selectors/ranking_based_response_selector/README.md index fae0d1b7b3..7773358a06 100644 --- a/response_selectors/ranking_based_response_selector/README.md +++ b/response_selectors/ranking_based_response_selector/README.md @@ -1,4 +1,3 @@ -<<<<<<< HEAD # Ranking-based Response Selector ## Description @@ -29,6 +28,9 @@ Parameter `FILTER_TOXIC_OR_BADLISTED` defines whether it filers out toxic hypoth text here **Output:** Ranking_based_response_selector service returns +- the selected skill’s name, +- the response text (which can be overwritten) +- the confidence level - the selected skill name, - the response text (which can be overwritten) - the confidence level @@ -38,4 +40,4 @@ text here ``` ## Dependencies ->>>>>>> readmes for dream multilingual dist +none From 889d8b101ae713bde5f50e650c87be6643f40fec Mon Sep 17 00:00:00 2001 From: nstsj Date: Wed, 25 Oct 2023 22:47:14 +0200 Subject: [PATCH 17/28] readme upd - rebasing the branch after pulling the fresh dev --- annotators/BadlistedWordsDetector/README.md | 3 ++- .../convers_evaluation_based_selector/README.md | 13 +++++++++++++ 2 files changed, 15 insertions(+), 1 deletion(-) diff --git a/annotators/BadlistedWordsDetector/README.md b/annotators/BadlistedWordsDetector/README.md index e5991f86ed..f81dc04652 100644 --- a/annotators/BadlistedWordsDetector/README.md +++ b/annotators/BadlistedWordsDetector/README.md @@ -4,7 +4,8 @@ Spacy-based user utterance annotator that detects words and phrases from the badlist ## I/O -**Input:** a list of user's uttetances +**Input:** a list of user's utterances [{"bad_words": True}, {"bad_words": False}, {"bad_words": True}] ## Dependencies +none \ No newline at end of file diff --git a/response_selectors/convers_evaluation_based_selector/README.md b/response_selectors/convers_evaluation_based_selector/README.md index 0e74ac3166..2d42c47c20 100644 --- a/response_selectors/convers_evaluation_based_selector/README.md +++ b/response_selectors/convers_evaluation_based_selector/README.md @@ -41,6 +41,19 @@ FALLBACK_FILE: a file name with fallbacks from `dream/common/fallbacks/` ## Input/Output **Input:** a list of hypotheses with their scores and metadata **Output:** a list of selected hypotheses with their scores and metadata +### How to run conversation evaluator locally + +`docker-compose -f docker-compose.yml -f dev.yml -f cpu.yml -f one_worker.yml up toxic_classification badlisted_words convers_evaluation_selector` + +Then use `--url`. + +Example of usage with url: ## Dependencies none +``` +python response_selectors/convers_evaluation_based_selector/measure_quality.py \ + --data_dir response_selectors/convers_evaluation_based_selector/labeled_data/ \ + --url http://0.0.0.0:8009/respond +``` +## Dependencies From 9762bdf790b9903f94ede0d5a25eb970ac0e40b6 Mon Sep 17 00:00:00 2001 From: nstsj Date: Wed, 25 Oct 2023 22:54:44 +0200 Subject: [PATCH 18/28] readme upd - rebasing the branch after pulling the fresh dev --- annotators/BadlistedWordsDetector/README.md | 3 +++ response_selectors/convers_evaluation_based_selector/README.md | 1 - 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/annotators/BadlistedWordsDetector/README.md b/annotators/BadlistedWordsDetector/README.md index f81dc04652..7777ea9669 100644 --- a/annotators/BadlistedWordsDetector/README.md +++ b/annotators/BadlistedWordsDetector/README.md @@ -1,6 +1,9 @@ # BadlistedWordsDetector ## Description +component_type: annotator +is_customizable: true + Spacy-based user utterance annotator that detects words and phrases from the badlist ## I/O diff --git a/response_selectors/convers_evaluation_based_selector/README.md b/response_selectors/convers_evaluation_based_selector/README.md index 2d42c47c20..2230b6ac7d 100644 --- a/response_selectors/convers_evaluation_based_selector/README.md +++ b/response_selectors/convers_evaluation_based_selector/README.md @@ -56,4 +56,3 @@ python response_selectors/convers_evaluation_based_selector/measure_quality.py \ --data_dir response_selectors/convers_evaluation_based_selector/labeled_data/ \ --url http://0.0.0.0:8009/respond ``` -## Dependencies From f0851abcbca7556bfd1b3a41e0ee5c2948acb023 Mon Sep 17 00:00:00 2001 From: nstsj Date: Wed, 25 Oct 2023 22:58:40 +0200 Subject: [PATCH 19/28] fixes during rebase --- annotators/BadlistedWordsDetector_ru/README.md | 2 ++ annotators/asr/README.md | 2 ++ 2 files changed, 4 insertions(+) diff --git a/annotators/BadlistedWordsDetector_ru/README.md b/annotators/BadlistedWordsDetector_ru/README.md index d2e32a5b0d..9113cd4f94 100644 --- a/annotators/BadlistedWordsDetector_ru/README.md +++ b/annotators/BadlistedWordsDetector_ru/README.md @@ -1,6 +1,8 @@ # BadlistedWordsDetector for Russian ## Description +component_type: annotator +is_customizable: true Spacy-based user utterance annotator that detects words and phrases from the badlist. This version of the annotator works for the Russian Language. diff --git a/annotators/asr/README.md b/annotators/asr/README.md index 4ce40764d3..e3803926f8 100644 --- a/annotators/asr/README.md +++ b/annotators/asr/README.md @@ -1,6 +1,8 @@ # ASR : Automatic Speech Recognition ## Description +component_type: annotator +is_customizable: true ASR component allows users to provide speech input via its `http://_service_name_:4343/asr?user_id=` endpoint. To do so, attach the recorded voice as a `.wav` file, 16KHz. From fee4410d24332a87f9b9b9f478f2e7edb3199465 Mon Sep 17 00:00:00 2001 From: nstsj Date: Wed, 25 Oct 2023 22:59:48 +0200 Subject: [PATCH 20/28] fixes during rebase --- annotators/ConversationEvaluator/README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/annotators/ConversationEvaluator/README.md b/annotators/ConversationEvaluator/README.md index 90767c4a2b..2c1b307223 100644 --- a/annotators/ConversationEvaluator/README.md +++ b/annotators/ConversationEvaluator/README.md @@ -19,4 +19,4 @@ tags with their probabilities ## Dependencies -none \ No newline at end of file +none From 11318c70977b04e32461aae45bbe74521cbb5595 Mon Sep 17 00:00:00 2001 From: nstsj Date: Wed, 25 Oct 2023 23:01:24 +0200 Subject: [PATCH 21/28] fixes during rebase --- annotators/asr/README.md | 4 ---- 1 file changed, 4 deletions(-) diff --git a/annotators/asr/README.md b/annotators/asr/README.md index e3803926f8..8cbc4e5e76 100644 --- a/annotators/asr/README.md +++ b/annotators/asr/README.md @@ -1,13 +1,9 @@ # ASR : Automatic Speech Recognition ## Description -component_type: annotator -is_customizable: true ASR component allows users to provide speech input via its `http://_service_name_:4343/asr?user_id=` endpoint. To do so, attach the recorded voice as a `.wav` file, 16KHz. -This component calculates overall ASR confidence for a given utterance and grades it as either *very low*, *low*, *medium*, *high* or *undefined* (for Amazon markup). - ## I/O **Input:** user utterance: recorded voice as a `.wav` file From 7fb04ff673f550577a57fdfb11fe45959d19332a Mon Sep 17 00:00:00 2001 From: nstsj Date: Wed, 25 Oct 2023 23:02:16 +0200 Subject: [PATCH 22/28] fixes during rebase --- annotators/BadlistedWordsDetector/README.md | 3 --- 1 file changed, 3 deletions(-) diff --git a/annotators/BadlistedWordsDetector/README.md b/annotators/BadlistedWordsDetector/README.md index 7777ea9669..f81dc04652 100644 --- a/annotators/BadlistedWordsDetector/README.md +++ b/annotators/BadlistedWordsDetector/README.md @@ -1,9 +1,6 @@ # BadlistedWordsDetector ## Description -component_type: annotator -is_customizable: true - Spacy-based user utterance annotator that detects words and phrases from the badlist ## I/O From 4ea75acf0ab10959c379efe26b0b6c138a5a9984 Mon Sep 17 00:00:00 2001 From: nstsj Date: Wed, 25 Oct 2023 23:03:08 +0200 Subject: [PATCH 23/28] fixes during rebase --- response_selectors/convers_evaluation_based_selector/README.md | 1 - 1 file changed, 1 deletion(-) diff --git a/response_selectors/convers_evaluation_based_selector/README.md b/response_selectors/convers_evaluation_based_selector/README.md index 2230b6ac7d..8e55e0601d 100644 --- a/response_selectors/convers_evaluation_based_selector/README.md +++ b/response_selectors/convers_evaluation_based_selector/README.md @@ -50,7 +50,6 @@ Then use `--url`. Example of usage with url: ## Dependencies -none ``` python response_selectors/convers_evaluation_based_selector/measure_quality.py \ --data_dir response_selectors/convers_evaluation_based_selector/labeled_data/ \ From 2429faaa5cb3c4e849d816c8ab69aa53e443628c Mon Sep 17 00:00:00 2001 From: nstsj Date: Mon, 28 Aug 2023 14:32:06 +0400 Subject: [PATCH 24/28] missing readme added --- annotators/summarization_annotator/README.md | 7 +++++++ 1 file changed, 7 insertions(+) create mode 100644 annotators/summarization_annotator/README.md diff --git a/annotators/summarization_annotator/README.md b/annotators/summarization_annotator/README.md new file mode 100644 index 0000000000..ba453f406b --- /dev/null +++ b/annotators/summarization_annotator/README.md @@ -0,0 +1,7 @@ +# Summarization Annotator +## Description + +## I/O + +## Dependencies +- annotators.spelling_preprocessing \ No newline at end of file From 4788051f32bc6f885876e264217b86f788536fcd Mon Sep 17 00:00:00 2001 From: nstsj Date: Wed, 25 Oct 2023 23:04:15 +0200 Subject: [PATCH 25/28] fixes during rebase --- annotators/property_extraction/README.md | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/annotators/property_extraction/README.md b/annotators/property_extraction/README.md index a2f651b55e..9ddba045b8 100644 --- a/annotators/property_extraction/README.md +++ b/annotators/property_extraction/README.md @@ -21,13 +21,19 @@ import requests utterances = [["I love going for a walk with my two dogs every day."], ["I like travelling in Italy with my husband. And you?"]] requests.post("http://0.0.0.0:8136/respond", json = {"utterances": utterances}).json() +``` ->>> [ +**Output example** +``` +[ {"triplets": [{"subject": "user", "relation": "like activity", "object": "walking"}, {"subject": "user", "relation": "have pet", "object": "two dogs"}]}, {"triplets": [{"subject": "user", "property": "marital status", "object": "husband"}, {"subject": "user", "relation": "like activity", "object": "travel"}]} ] ``` +## Dependencies +none + **Output** ## Dependencies \ No newline at end of file From 97956c483881668894f161a24f9cc191b5adec8f Mon Sep 17 00:00:00 2001 From: nstsj Date: Tue, 28 Nov 2023 03:55:36 +0100 Subject: [PATCH 26/28] fixed merge conflicts, re-updated files --- annotators/BadlistedWordsDetector/README.md | 12 ++- .../BadlistedWordsDetector_ru/README.md | 19 +++- annotators/COMeT/README.md | 101 ++++++++++++++++-- .../DeepPavlovEmotionClassification/README.md | 4 +- annotators/NER/README.md | 21 +++- annotators/custom_entity_linking/README.md | 20 +++- annotators/property_extraction/README.md | 20 ++-- .../README.md | 56 ++++++++-- .../ranking_based_response_selector/README.md | 72 ++++++++----- 9 files changed, 258 insertions(+), 67 deletions(-) diff --git a/annotators/BadlistedWordsDetector/README.md b/annotators/BadlistedWordsDetector/README.md index f81dc04652..b46cb2680c 100644 --- a/annotators/BadlistedWordsDetector/README.md +++ b/annotators/BadlistedWordsDetector/README.md @@ -4,8 +4,16 @@ Spacy-based user utterance annotator that detects words and phrases from the badlist ## I/O -**Input:** a list of user's utterances -[{"bad_words": True}, {"bad_words": False}, {"bad_words": True}] +**Input:** a list of user's utterances +``` +["fucking hell", "he mishit the shot", "you asshole"] +``` + +**Output:** words and their tags +``` + [{"bad_words": True}, {"bad_words": False}, {"bad_words": True}] +``` + ## Dependencies none \ No newline at end of file diff --git a/annotators/BadlistedWordsDetector_ru/README.md b/annotators/BadlistedWordsDetector_ru/README.md index 9113cd4f94..7ea8f487cf 100644 --- a/annotators/BadlistedWordsDetector_ru/README.md +++ b/annotators/BadlistedWordsDetector_ru/README.md @@ -1,14 +1,23 @@ # BadlistedWordsDetector for Russian ## Description -component_type: annotator -is_customizable: true -Spacy-based user utterance annotator that detects words and phrases from the badlist. This version of the annotator works for the Russian Language. +Spacy-based user utterance annotator that detects words and phrases from the badlist. + +This version of the annotator works for the Russian Language. ## I/O -**Input:** a list of user's uttetances -**Output:** words and their tags (`{"bad_words": False}` or `{"bad_words": True}`) +**Input:** +Takes a list of user's utterances +``` +["не пизди.", "застрахуйте уже его", "пошел нахер!"] +``` + +**Output:** +Returns words and their tags +``` +[{"bad_words": True}, {"bad_words": False}, {"bad_words": True}] +``` ## Dependencies none \ No newline at end of file diff --git a/annotators/COMeT/README.md b/annotators/COMeT/README.md index 41034846c3..213b9417ca 100644 --- a/annotators/COMeT/README.md +++ b/annotators/COMeT/README.md @@ -39,18 +39,103 @@ docker-compose -f docker-compose.yml -f local.yml exec comet-conceptnet bash tes ## Input/Output -**Input** +**Input** - hypotheses: possible assistant's replies - currentUtterance: latest reply from a user -- pastResponses: a list of user's utterances +- pastResponses: a list of user's utterances +an input example (): +``` +{ + "input": "PersonX went to a mall", + "category": [ + "xReact", + "xNeed", + "xAttr", + "xWant", + "oEffect", + "xIntent", + "oReact" + ] +} + +``` **Output** -a list of probabilities about the utterance: -- isResponseComprehensible -- isResponseErroneous -- isResponseInteresting -- isResponseOnTopic -- responseEngagesUser +a list of probabilities about the utterance based on categories: +- xReact +- xNeed +- xAttr +- xWant +- oEffect +- xIntent +- oReact + +an output example (): +``` + "xReact": { + "beams": [ + "satisfied", + "happy", + "excited" + ], + "effect_type": "xReact", + "event": "PersonX went to a mall" + }, + "xNeed": { + "beams": [ + "to drive to the mall", + "to get in the car", + "to drive to the mall" + ], + "effect_type": "xNeed", + "event": "PersonX went to a mall" + }, + "xAttr": { + "beams": [ + "curious", + "fashionable", + "interested" + ], + "effect_type": "xAttr", + "event": "PersonX went to a mall" + }, + "xWant": { + "beams": [ + "to buy something", + "to go home", + "to shop" + ], + "effect_type": "xWant", + "event": "PersonX went to a mall" + }, + "oEffect": { + "beams": [ + "they go to the store", + "they go to the mall" + ], + "effect_type": "oEffect", + "event": "PersonX went to a mall" + }, + "xIntent": { + "beams": [ + "to buy something", + "to shop", + "to buy things" + ], + "effect_type": "xIntent", + "event": "PersonX went to a mall" + }, + "oReact": { + "beams": [ + "happy", + "interested" + ], + "effect_type": "oReact", + "event": "PersonX went to a mall" + } +} +``` + ## Dependencies diff --git a/annotators/DeepPavlovEmotionClassification/README.md b/annotators/DeepPavlovEmotionClassification/README.md index 1957519484..79571b1a06 100644 --- a/annotators/DeepPavlovEmotionClassification/README.md +++ b/annotators/DeepPavlovEmotionClassification/README.md @@ -2,11 +2,11 @@ ## Description -BERT Base model for emotion classification which learned at the custom dataset (described more precisely in (our article)[link here]) +BERT Base model for emotion classification ## I/O -**Inpunt** +**Input** **Output:** diff --git a/annotators/NER/README.md b/annotators/NER/README.md index 9a5a832efd..f9bc5c824d 100644 --- a/annotators/NER/README.md +++ b/annotators/NER/README.md @@ -1,9 +1,28 @@ # Title +Named Entity Recognition Annotator + ## Description +Extracts people names, locations and names of organizations from an uncased text ## Input/Output **Input** +A list of user utterances +``` +["john peterson is my brother.", "he lives in New York."] +``` + + **Output** +A user utterance annotated by +- confidence level +- named entity's position in a sentence (`start_pos` and `end_pos`) +- the named the entity itself +- the named entity type + +``` + [{"confidence": 1, "end_pos": 5, "start_pos": 3, "text": "New York", "type": "LOC"}], +``` -## Dependencies \ No newline at end of file +## Dependencies +none \ No newline at end of file diff --git a/annotators/custom_entity_linking/README.md b/annotators/custom_entity_linking/README.md index 759da898f3..50ae66a6e3 100644 --- a/annotators/custom_entity_linking/README.md +++ b/annotators/custom_entity_linking/README.md @@ -19,15 +19,25 @@ Relation examples: ## I/O -**Inpunt** -user_id, entity substring, entity_tags +**Input** +Takes a list of user_id, entity substring, entity_tags -**Output:** +An input example: +``` +``` + +**Output:** processed information about: - entities - entity_id (ids for multiple entities) - entity_confidence score - entity_id_tags - + +An output example: +``` +``` + ## Dependencies - annotators: `NER`, `entity_detection` and `spacy_nounphrases` +- annotators.ner +- annotators.entity_detection +- annotators.spacy_nounphrases diff --git a/annotators/property_extraction/README.md b/annotators/property_extraction/README.md index 9ddba045b8..c88f644cd6 100644 --- a/annotators/property_extraction/README.md +++ b/annotators/property_extraction/README.md @@ -14,16 +14,23 @@ Entity generator - a se2seq model which generates the subject and object for eac ## I/O -**Input example** +**Input** +Takes user utterances as a list of lists +An input example: ```python import requests -utterances = [["I love going for a walk with my two dogs every day."], ["I like travelling in Italy with my husband. And you?"]] +utterances = [["I love going for a walk with my two dogs every day."], + ["I like travelling in Italy with my husband. And you?"]] + requests.post("http://0.0.0.0:8136/respond", json = {"utterances": utterances}).json() ``` -**Output example** +**Output** +Returns annotated triplets *subject-relation-object* with information about entities and their relations + +An output example: ``` [ {"triplets": [{"subject": "user", "relation": "like activity", "object": "walking"}, {"subject": "user", "relation": "have pet", "object": "two dogs"}]}, @@ -32,8 +39,5 @@ requests.post("http://0.0.0.0:8136/respond", json = {"utterances": utterances}). ``` ## Dependencies -none - -**Output** - -## Dependencies \ No newline at end of file +- annotators.ner +- annotators.entity_linking diff --git a/response_selectors/convers_evaluation_based_selector/README.md b/response_selectors/convers_evaluation_based_selector/README.md index 8e55e0601d..7479494641 100644 --- a/response_selectors/convers_evaluation_based_selector/README.md +++ b/response_selectors/convers_evaluation_based_selector/README.md @@ -5,14 +5,14 @@ Response Selector is a component selecting final response among the given hypotheses by different skills. The Tag- and Evaluation--based Response Selector utilizes a complicated approach which aims to -prioritize scripted skills while having an opportunity to provide a system-initiative via so-called linking questions that bring conversation to the scripts. +prioritize scripted skills while having an opportunity to provide a system-initiative via so-called linking questions that bring conversation to the scripts. A final hypothesis could be a combination of a hypothesis and linking question. The approach is most suitable for distributions where the most of the responses are implied to be by scripts. ### Parameters -The algorithm contains a large number of parameters which control the filtration and prioritization rules. +The algorithm contains a large number of parameters which control the filtration and prioritization rules. The algorithm filers out toxic hypotheses. ``` @@ -39,19 +39,53 @@ FALLBACK_FILE: a file name with fallbacks from `dream/common/fallbacks/` ``` ## Input/Output -**Input:** a list of hypotheses with their scores and metadata -**Output:** a list of selected hypotheses with their scores and metadata +**Input:** a list of hypotheses generated by corresponding skills, with their scores and metadata +**Output:** the final hypothesis chosen as a reply + +A partial example of such a response selector's output: +``` +{ + "skill_name": "program_y", + "annotations": { + "toxic_classification": { + "identity_hate": 8.749961853027344e-05, + "insult": 0.00024232268333435059, + "obscene": 2.828240394592285e-05, + "severe_toxic": 1.8358230590820312e-05, + "sexual_explicit": 2.9712915420532227e-05, + "threat": 6.490945816040039e-05, + "toxic": 0.00043845176696777344 + }, + "stop_detect": { + "stop": 0.5808720588684082, + "continue": 0.45234695076942444 + }, + "convers_evaluator_annotator": { + "isResponseComprehensible": 0.984, + "isResponseErroneous": 0.614, + "isResponseInteresting": 0.253, + "isResponseOnTopic": 0.226, + "responseEngagesUser": 0.56 + }, + "badlisted_words": { + "inappropriate": false, + "profanity": false, + "restricted_topics": false + } + }, + "text": "Good Morning, this is an Alexa Prize Socialbot! How are you?", + "confidence": 0.98 + } + +``` + + ### How to run conversation evaluator locally `docker-compose -f docker-compose.yml -f dev.yml -f cpu.yml -f one_worker.yml up toxic_classification badlisted_words convers_evaluation_selector` Then use `--url`. -Example of usage with url: - ## Dependencies -``` -python response_selectors/convers_evaluation_based_selector/measure_quality.py \ - --data_dir response_selectors/convers_evaluation_based_selector/labeled_data/ \ - --url http://0.0.0.0:8009/respond -``` +none + diff --git a/response_selectors/ranking_based_response_selector/README.md b/response_selectors/ranking_based_response_selector/README.md index 7773358a06..6dd6b15706 100644 --- a/response_selectors/ranking_based_response_selector/README.md +++ b/response_selectors/ranking_based_response_selector/README.md @@ -8,36 +8,58 @@ to rank hypotheses and selects the best ranked one. ### Parameters -Utilizes annotations by `SENTENCE_RANKER_ANNOTATION_NAME` candidate annotator. -In case of absence of these annotations, utilizes provided `SENTENCE_RANKER_SERVICE_URL` to annotate hypotheses +Utilizes annotations by `SENTENCE_RANKER_ANNOTATION_NAME` candidate annotator. +In case of absence of these annotations, utilizes provided `SENTENCE_RANKER_SERVICE_URL` to annotate hypotheses according to `N_UTTERANCES_CONTEXT` last utterances. Parameter `FILTER_TOXIC_OR_BADLISTED` defines whether it filers out toxic hypotheses or not. -## Dependencies - -- either candidate annotations by `SENTENCE_RANKER_ANNOTATION_NAME` or service `SENTENCE_RANKER_SERVICE_URL`. -======= -# Ranking_based_response_selector - -## Description - +**Output:** +Ranking_based_response_selector service returns ++ the selected skill’s name, ++ the response text (which can be overwritten) ++ the confidence level ++ the selected skill name, ++ the response text (which can be overwritten) ++ the confidence level -## Input/Output - -**Input:** -text here - -**Output:** Ranking_based_response_selector service returns -- the selected skill’s name, -- the response text (which can be overwritten) -- the confidence level -- the selected skill name, -- the response text (which can be overwritten) -- the confidence level - -``` +A partial example of such a response selector's output: ``` +{ + "skill_name": "movie_tfidf_retrieval", + "annotations": { + "toxic_classification": { + "identity_hate": 0.0001259446144104004, + "insult": 0.00027686357498168945, + "obscene": 5.97834587097168e-05, + "severe_toxic": 3.403425216674805e-05, + "sexual_explicit": 8.13603401184082e-05, + "threat": 0.00012931227684020996, + "toxic": 0.0005629658699035645 + }, + "stop_detect": { + "stop": 0.5833511352539062, + "continue": 0.46003755927085876 + }, + "convers_evaluator_annotator": { + "isResponseComprehensible": 0.281, + "isResponseErroneous": 0.531, + "isResponseInteresting": 0.228, + "isResponseOnTopic": 0.254, + "responseEngagesUser": 0.536 + }, + "badlisted_words": { + "inappropriate": false, + "profanity": false, + "restricted_topics": false + } + }, + "text": "i got you haha. what do you think about celebrities? judge judy makes 123, 000 per episode apparently!", + "confidence": 0.38232852805460565 + }, +``` ## Dependencies -none + +- either candidate annotations by `SENTENCE_RANKER_ANNOTATION_NAME` or service `SENTENCE_RANKER_SERVICE_URL`. + From 58bf98b7fe2c62bb27ee8137f3ee5ea42819635e Mon Sep 17 00:00:00 2001 From: nstsj Date: Tue, 28 Nov 2023 05:05:13 +0100 Subject: [PATCH 27/28] ancient components readmes upd: added more content and examples --- skills/dummy_skill/README.md | 15 ++++++++---- skills/dummy_skill_dialog/README.md | 13 ++++++++-- skills/personal_info_skill/README.md | 17 ++++++++++--- skills/personality_catcher/README.md | 13 ++++++---- skills/program_y_deepy/README.md | 3 ++- skills/retrieval_chitchat/README.md | 2 +- skills/small_talk_skill/README.md | 24 ++++++++++++++++--- skills/superbowl_skill/README.md | 12 ++++++++-- .../topicalchat_convert_retrieval/README.md | 3 ++- skills/topicalchat_tfidf_retrieval/README.md | 3 ++- skills/valentines_day_skill/README.md | 13 ++++++++-- 11 files changed, 93 insertions(+), 25 deletions(-) diff --git a/skills/dummy_skill/README.md b/skills/dummy_skill/README.md index 53cec85470..b0ff29d704 100644 --- a/skills/dummy_skill/README.md +++ b/skills/dummy_skill/README.md @@ -2,14 +2,21 @@ ## Description - -Russian Random questions are collected from https://mensby.com/women/relations/150-voprosov-chtoby-luchshe-uznat-sobesednika-ili-sobesednicu -and https://habr.com/ru/company/testutor/blog/298180/ +TBA # I/O **Input** +A user's utterance +``` + +``` **Output** -## Dependencies +``` + +``` + +## Dependencies +none diff --git a/skills/dummy_skill_dialog/README.md b/skills/dummy_skill_dialog/README.md index 9a5a832efd..8b70508920 100644 --- a/skills/dummy_skill_dialog/README.md +++ b/skills/dummy_skill_dialog/README.md @@ -1,9 +1,18 @@ -# Title +# Dummy Dialogue Skill ## Description +A fallback skill that is activated when the bot failed to understand the user's utterance. ## Input/Output **Input** +A user utterance + +`["Will intelligent computers take over the world?"]` + **Output** -## Dependencies \ No newline at end of file +`["I really do not know what to answer."]` + + +## Dependencies +none \ No newline at end of file diff --git a/skills/personal_info_skill/README.md b/skills/personal_info_skill/README.md index 9a5a832efd..5ec935b2c8 100644 --- a/skills/personal_info_skill/README.md +++ b/skills/personal_info_skill/README.md @@ -1,9 +1,20 @@ -# Title +# Personal Info Skill ## Description - +A skill that is able to parse user's input and retrieve person's name. ## Input/Output **Input** +A user utterance +``` +["my name is john."] +``` **Output** +A parsed utterance, retrieved name and a reply + +``` + [["Nice to meet you, John."], + ["text": "john", "type": "PER"]] +``` -## Dependencies \ No newline at end of file +## Dependencies +none \ No newline at end of file diff --git a/skills/personality_catcher/README.md b/skills/personality_catcher/README.md index 9778349dc5..d2894bb520 100644 --- a/skills/personality_catcher/README.md +++ b/skills/personality_catcher/README.md @@ -1,15 +1,18 @@ +# Personality Catcher Skill ## Description -Этот скил сделан для смены персоны бота. -В решении много костылей, поэтому решение временное. -По функционалу должно использоваться для только разработчиками. +This skill is created in order to be able to change the bot personality. +Note: this is a paragon of a crutch-driven development, so the solution temporal (and only for devs) ## Getting started -Для смены персоны необходимо отправить через телеграмм сообщение `/start` - для создания нового диалога. После вместе с командой `/new_persona` отравить персону через знак новой строки `\n`. Знаком новой строки `\n` разделяются предложения из персоны. +To change the bot persona: +1. in Telegram, when talking to your bot, send `/start` command +2. Then using a command `/new_persona` send the new persona description (use `\n` at the end of the sentences) +3.`\n` works as a separator for sentences. ## Crutches -Все костыли этого решения помечены записью `# TODO: rm crutch of personality_catcher` +All the dev crutches are marked as `# TODO: rm crutch of personality_catcher` diff --git a/skills/program_y_deepy/README.md b/skills/program_y_deepy/README.md index 9a5a832efd..1879e16ca6 100644 --- a/skills/program_y_deepy/README.md +++ b/skills/program_y_deepy/README.md @@ -1,6 +1,7 @@ -# Title +# Program-Y Deepy Skill ## Description + ## Input/Output **Input** diff --git a/skills/retrieval_chitchat/README.md b/skills/retrieval_chitchat/README.md index 9a5a832efd..da75c118b3 100644 --- a/skills/retrieval_chitchat/README.md +++ b/skills/retrieval_chitchat/README.md @@ -1,4 +1,4 @@ -# Title +# Retrieval ChitChat Skill ## Description ## Input/Output diff --git a/skills/small_talk_skill/README.md b/skills/small_talk_skill/README.md index 9a5a832efd..be16be6680 100644 --- a/skills/small_talk_skill/README.md +++ b/skills/small_talk_skill/README.md @@ -1,9 +1,27 @@ -# Title -## Description +# Small Talk Skill +An example of DFF-template-based skill, this skill is able to reply to queries that look like a small talk. +This skill is able to respond to user utterances about: +- life +- love and friendship +- food +- movies +- news +- politics +- weather +- books +- etc ## Input/Output **Input** +A user utterance +```["Do you like movies about superheroes?"]``` + **Output** +a reply from a set of pre-made keyword-relevant phrases + +```["I like them, too. I've heard that many people love superheroes. Do you have a favorite one?"]``` -## Dependencies \ No newline at end of file +More on how to create skills like this one [here](https://docs.dream.deeppavlov.ai/guides/custom_skill) +## Dependencies +none \ No newline at end of file diff --git a/skills/superbowl_skill/README.md b/skills/superbowl_skill/README.md index 9a5a832efd..48569cf364 100644 --- a/skills/superbowl_skill/README.md +++ b/skills/superbowl_skill/README.md @@ -1,9 +1,17 @@ -# Title +# DFF Valentine's Day Skill ## Description +An example of DFF-template-based skill, this skill is able to reply to queries about Superbowl. ## Input/Output **Input** +A user utterance (Superbowl-related since the skill is topical) +```["tell me a fun fact about the super bowl"]``` + **Output** +a reply from a set of pre-made keyword-relevant phrases +```["Here is the fun fact about super bowl from kidzworld.com: 8 million pounds of guacamole is consumed on Super Bowl Sunday"]``` -## Dependencies \ No newline at end of file +More on how to create skills like this one [here](https://docs.dream.deeppavlov.ai/guides/custom_skill) +## Dependencies +none \ No newline at end of file diff --git a/skills/topicalchat_convert_retrieval/README.md b/skills/topicalchat_convert_retrieval/README.md index 9a5a832efd..6b08fbe075 100644 --- a/skills/topicalchat_convert_retrieval/README.md +++ b/skills/topicalchat_convert_retrieval/README.md @@ -1,6 +1,7 @@ -# Title +# Topical Convert Retrieval Skill ## Description + ## Input/Output **Input** diff --git a/skills/topicalchat_tfidf_retrieval/README.md b/skills/topicalchat_tfidf_retrieval/README.md index 9a5a832efd..0b906b90e9 100644 --- a/skills/topicalchat_tfidf_retrieval/README.md +++ b/skills/topicalchat_tfidf_retrieval/README.md @@ -1,6 +1,7 @@ -# Title +# Topical Chat TF-IDF Retrieval Skill ## Description + ## Input/Output **Input** diff --git a/skills/valentines_day_skill/README.md b/skills/valentines_day_skill/README.md index 9a5a832efd..a201893761 100644 --- a/skills/valentines_day_skill/README.md +++ b/skills/valentines_day_skill/README.md @@ -1,9 +1,18 @@ -# Title +# DFF Valentine's Day Skill ## Description +An example of DFF-template-based skill, this skill is able to reply to queries about love. ## Input/Output **Input** +A user utterance (love-related since the skill is topical) +```["tell me about the love"]``` + **Output** +a reply from a set of pre-made keyword-relevant phrases + +```"Love is a friendship set to music. Quote by Joseph Campbell"``` -## Dependencies \ No newline at end of file +More on how to create skills like this one [here](https://docs.dream.deeppavlov.ai/guides/custom_skill) +## Dependencies +none \ No newline at end of file From 8fc38c9bb65588b59aee1474a3e4f8f4037ad937 Mon Sep 17 00:00:00 2001 From: nstsj Date: Tue, 28 Nov 2023 06:05:17 +0100 Subject: [PATCH 28/28] added readmes for main Dream distributions, explaining their purpose and components --- assistant_dists/dream_emotion/README.md | 25 +++++++++++++ assistant_dists/dream_kg/README.md | 29 +++++++++++++++ assistant_dists/dream_kg_prompted/README.md | 29 +++++++++++++++ assistant_dists/dream_russian/README.md | 37 +++++++++++++++++++ .../multiskill_ai_assistant/README.md | 24 ++++++++++++ skills/dff_wiki_skill/README.md | 2 +- 6 files changed, 145 insertions(+), 1 deletion(-) create mode 100644 assistant_dists/dream_emotion/README.md create mode 100644 assistant_dists/dream_kg/README.md create mode 100644 assistant_dists/dream_kg_prompted/README.md create mode 100644 assistant_dists/dream_russian/README.md create mode 100644 assistant_dists/multiskill_ai_assistant/README.md diff --git a/assistant_dists/dream_emotion/README.md b/assistant_dists/dream_emotion/README.md new file mode 100644 index 0000000000..5da8c14723 --- /dev/null +++ b/assistant_dists/dream_emotion/README.md @@ -0,0 +1,25 @@ +# DeepPavlov Dream Emotion distribution + +Distribution is a set of configuration files that define which components are used in this particular assistant. +This defines an assistant itself. + +*(different Dream distributions may use different Annotators and Response Selectors, feature various Skills, and even speak different languages)* +- see how DeepPavlov Dream distribution works [here](https://docs.dream.deeppavlov.ai/dream_scheme.png) +- for full list of available distributions, click [here](https://docs.dream.deeppavlov.ai/ref_materials/distributions) + +## Description +This particular DeepPavlov Dream distribution focuses on a task of emotion classification during the dialogue, and contains a mix of Generative and DFF-template-based Skills. + + +## Services used in this DeepPavlov Dream Distribution +- sentseg +- combined-classification +- sentence-ranker +- prompt-selector +- openai-api-chatgpt +- dff-dream-persona-chatgpt-prompted-skill +- dff-dream-faq-prompted-skill +- openai-api-chatgpt-16k +- bot-emotion-classifier +- emotional-bot-response +- emotion-ranking-based-response-selector \ No newline at end of file diff --git a/assistant_dists/dream_kg/README.md b/assistant_dists/dream_kg/README.md new file mode 100644 index 0000000000..3bae2e4845 --- /dev/null +++ b/assistant_dists/dream_kg/README.md @@ -0,0 +1,29 @@ +# DeepPavlov Dream KG distribution + +Distribution is a set of configuration files that define which components are used in this particular assistant. +This defines an assistant itself. + +*(different Dream distributions may use different Annotators and Response Selectors, feature various Skills, and even speak different languages)* +- see how DeepPavlov Dream distribution works [here](https://docs.dream.deeppavlov.ai/dream_scheme.png) +- for full list of available distributions, click [here](https://docs.dream.deeppavlov.ai/ref_materials/distributions) + +## Description +This particular DeepPavlov Dream distribution is powered by [Custom Knowledge Graph](https://docs.dream.deeppavlov.ai/ref_materials/custom_kg) option. +It is a tool for collecting, organising and storing data that can be used in any Dream Skill to offer the user a personalized experience and make bot replies more relevant. + + +## Services used in this DeepPavlov Dream Distribution +- sentence-ranker +- sentseg +- ranking-based-response-selector +- ner +- entity-linking +- combined-classification +- entity-detection +- property-extraction +- custom-entity-linking +- terminusdb-server +- user-knowledge-memorizer +- dff-user-kg-skill +- dff-travel-italy-skill + diff --git a/assistant_dists/dream_kg_prompted/README.md b/assistant_dists/dream_kg_prompted/README.md new file mode 100644 index 0000000000..e4ef5f86ab --- /dev/null +++ b/assistant_dists/dream_kg_prompted/README.md @@ -0,0 +1,29 @@ +# DeepPavlov Dream KG-prompted distribution + +Distribution is a set of configuration files that define which components are used in this particular assistant. +This defines an assistant itself. + +*(different Dream distributions may use different Annotators and Response Selectors, feature various Skills, and even speak different languages)* +- see how DeepPavlov Dream distribution works [here](https://docs.dream.deeppavlov.ai/dream_scheme.png) +- for full list of available distributions, click [here](https://docs.dream.deeppavlov.ai/ref_materials/distributions) + +## Description +This particular DeepPavlov Dream distribution is powered by [Custom Knowledge Graph](https://docs.dream.deeppavlov.ai/ref_materials/custom_kg) option. +It is a tool for collecting, organising and storing data that can be used in any Dream Skill to offer the user a personalized experience and make bot replies more relevant. +In addition, this distribution contains a Generative Skill utilising OpenAI API. + +## Services used in this DeepPavlov Dream Distribution +- sentseg +- ranking-based-response-selector +- ner +- entity-linking +- combined-classification +- entity-detection +- terminusdb-server +-user-knowledge-memorizer +- property-extraction +- custom-entity-linking +- dff-knowledge-prompted-skill +- sentence-ranker +- prompt-selector +- openai-api-chatgpt \ No newline at end of file diff --git a/assistant_dists/dream_russian/README.md b/assistant_dists/dream_russian/README.md new file mode 100644 index 0000000000..528fe8d3d0 --- /dev/null +++ b/assistant_dists/dream_russian/README.md @@ -0,0 +1,37 @@ +# DeepPavlov Dream the Russian Language distribution + +Distribution is a set of configuration files that define which components are used in this particular assistant. +This defines an assistant itself. + +*(different Dream distributions may use different Annotators and Response Selectors, feature various Skills, and even speak different languages)* +- see how DeepPavlov Dream distribution works [here](https://docs.dream.deeppavlov.ai/dream_scheme.png) +- for full list of available distributions, click [here](https://docs.dream.deeppavlov.ai/ref_materials/distributions) + +## Description +This particular DeepPavlov Dream distribution is adapted for the Russian Language. +This is the same distribution as main Dream dist, but with NLG/NLU models trained to work for texts in Russian. + +## Services used in this DeepPavlov Dream Distribution +- dff-program-y-ru-skill +- convers-evaluation-selector-ru +- dff-intent-responder-ru-skill +- intent-catcher-ru +- badlisted-words-ru +- ner-ru +- personal-info-ru-skill +- sentseg-ru +- spelling-preprocessing-ru +- entity-linking-ru +- wiki-parser-ru +- dff-generative-ru-skill +- dff-friendship-ru-skill +- entity-detection-ru +- dialogpt-ru +- dff-template-skill +- spacy-annotator-ru +- dialogrpt-ru +- fact-retrieval-ru +- text-qa-ru +- summarization-annotator +- rut5-summarizer +- combined-classification-ru \ No newline at end of file diff --git a/assistant_dists/multiskill_ai_assistant/README.md b/assistant_dists/multiskill_ai_assistant/README.md new file mode 100644 index 0000000000..f45ce31652 --- /dev/null +++ b/assistant_dists/multiskill_ai_assistant/README.md @@ -0,0 +1,24 @@ +# DeepPavlov Dream MultiSkill-AI distribution + +Distribution is a set of configuration files that define which components are used in this particular assistant. +This defines an assistant itself. + +*(different Dream distributions may use different Annotators and Response Selectors, feature various Skills, and even speak different languages)* +- see how DeepPavlov Dream distribution works [here](https://docs.dream.deeppavlov.ai/dream_scheme.png) +- for full list of available distributions, click [here](https://docs.dream.deeppavlov.ai/ref_materials/distributions) + +## Description +This DeepPavlov Dream distribution contains both DFF-template-based and Generative skills + +## Services used in this DeepPavlov Dream Distribution +- sentseg +- llm-based-response-selector +- combined-classification +- sentence-ranker +- prompt-selector +- openai-api-chatgpt +- dff-dream-persona-chatgpt-prompted-skill +- dff-casual-email-prompted-skill +- dff-meeting-notes-prompted-skill +- dff-official-email-prompted-skill +- dff-plan-for-article-prompted-skill \ No newline at end of file diff --git a/skills/dff_wiki_skill/README.md b/skills/dff_wiki_skill/README.md index 42e6bc1e9f..bf6d42715a 100644 --- a/skills/dff_wiki_skill/README.md +++ b/skills/dff_wiki_skill/README.md @@ -5,7 +5,7 @@ Template has dialog flows based on programy (`repeating`) and based on valila py -# Importan changes in files of the agent +# Important changes in files of the agent docker-compose.yml ```yml dff-template: