From 6acaa390b93ce99c9f34f8b0fd0489825497d2f2 Mon Sep 17 00:00:00 2001 From: Eric Smith Date: Tue, 22 Dec 2020 13:37:57 -0500 Subject: [PATCH] Move README + add refs (#3319) --- parlai/crowdsourcing/{tasks => }/README.md | 0 parlai/crowdsourcing/tasks/acute_eval/README.md | 2 ++ parlai/crowdsourcing/tasks/chat_demo/README.md | 2 ++ parlai/crowdsourcing/tasks/fast_acute/README.md | 2 ++ parlai/crowdsourcing/tasks/turn_annotations/README.md | 2 ++ parlai/crowdsourcing/tasks/turn_annotations_static/README.md | 2 ++ 6 files changed, 10 insertions(+) rename parlai/crowdsourcing/{tasks => }/README.md (100%) diff --git a/parlai/crowdsourcing/tasks/README.md b/parlai/crowdsourcing/README.md similarity index 100% rename from parlai/crowdsourcing/tasks/README.md rename to parlai/crowdsourcing/README.md diff --git a/parlai/crowdsourcing/tasks/acute_eval/README.md b/parlai/crowdsourcing/tasks/acute_eval/README.md index df9477b395e..d9cc9a657bd 100644 --- a/parlai/crowdsourcing/tasks/acute_eval/README.md +++ b/parlai/crowdsourcing/tasks/acute_eval/README.md @@ -27,6 +27,8 @@ The `run.py` script is designed to allow you to run this entire task from comman python parlai/crowdsourcing/tasks/acute_eval/run.py \ mephisto.blueprint.pairings_filepath=${REPO_FOLDER}/parlai/crowdsourcing/tasks/acute_eval/pairings.jsonl +**NOTE**: See [parlai/crowdsourcing/README.md](https://github.com/facebookresearch/ParlAI/blob/master/parlai/crowdsourcing/README.md) for general tips on running `parlai.crowdsourcing` tasks, such as how to specify your own YAML file of configuration settings, how to run tasks live, how to set parameters on the command line, etc. + ## Formatting conversation data This task code assumes that you've parsed and saved your collected conversations in a simple .jsonl format. The path to this file should be passed in as `mephisto.blueprint.pairings_filepath=${PATH_TO_FILE}`. diff --git a/parlai/crowdsourcing/tasks/chat_demo/README.md b/parlai/crowdsourcing/tasks/chat_demo/README.md index 665238d7954..18c0dd8a6c5 100644 --- a/parlai/crowdsourcing/tasks/chat_demo/README.md +++ b/parlai/crowdsourcing/tasks/chat_demo/README.md @@ -3,3 +3,5 @@ Chat task using two ParlAI-style agents. Code can be extended to change the number of human agents or to add bot agents. See the Mephisto [README](https://github.com/facebookresearch/Mephisto/blob/master/examples/parlai_chat_task_demo/README.md) for more details. Since this task is imported from the Mephisto repo, `run.py` will look in `${MEPHISTO_REPO}/examples/parlai_chat_task_demo/conf/` for YAML configuration files by default. See [here](https://github.com/facebookresearch/ParlAI/tree/master/parlai/crowdsourcing/tasks#specifying-your-own-yaml-file) for how to specify a custom folder for configuration files. + +**NOTE**: See [parlai/crowdsourcing/README.md](https://github.com/facebookresearch/ParlAI/blob/master/parlai/crowdsourcing/README.md) for general tips on running `parlai.crowdsourcing` tasks, such as how to specify your own YAML file of configuration settings, how to run tasks live, how to set parameters on the command line, etc. diff --git a/parlai/crowdsourcing/tasks/fast_acute/README.md b/parlai/crowdsourcing/tasks/fast_acute/README.md index 089d3772f46..502e3ede47d 100644 --- a/parlai/crowdsourcing/tasks/fast_acute/README.md +++ b/parlai/crowdsourcing/tasks/fast_acute/README.md @@ -8,6 +8,8 @@ The scripts in this directory will allow you to run all the steps of [ACUTE-Eval Both types are discussed below. +**NOTE**: See [parlai/crowdsourcing/README.md](https://github.com/facebookresearch/ParlAI/blob/master/parlai/crowdsourcing/README.md) for general tips on running `parlai.crowdsourcing` tasks, such as how to specify your own YAML file of configuration settings, how to run tasks live, how to set parameters on the command line, etc. + ## How to run Fast ACUTE if you need to produce model self-chats ### 1. Choose the self-chat task diff --git a/parlai/crowdsourcing/tasks/turn_annotations/README.md b/parlai/crowdsourcing/tasks/turn_annotations/README.md index 9d08e9985d6..628181f6c3d 100644 --- a/parlai/crowdsourcing/tasks/turn_annotations/README.md +++ b/parlai/crowdsourcing/tasks/turn_annotations/README.md @@ -2,6 +2,8 @@ This task will collect conversations between a human and a model. After each response by the model, the human will be prompted to annotate the model's response by selecting checkboxes that represent customizable attributes of that response. +**NOTE**: See [parlai/crowdsourcing/README.md](https://github.com/facebookresearch/ParlAI/blob/master/parlai/crowdsourcing/README.md) for general tips on running `parlai.crowdsourcing` tasks, such as how to specify your own YAML file of configuration settings, how to run tasks live, how to set parameters on the command line, etc. + ## Launching Call `run.py` to run this task with the default parameters, as set by `conf/example.yaml`. diff --git a/parlai/crowdsourcing/tasks/turn_annotations_static/README.md b/parlai/crowdsourcing/tasks/turn_annotations_static/README.md index 48f0ed33cc7..9057267cdd7 100644 --- a/parlai/crowdsourcing/tasks/turn_annotations_static/README.md +++ b/parlai/crowdsourcing/tasks/turn_annotations_static/README.md @@ -12,3 +12,5 @@ Two variants of the blueprint are supported: For both variants of the blueprint, it is required to pass in your own file of conversations with `mephisto.blueprint.data_jsonl=${PATH_TO_CONVERSATIONS}`. See `turn_annotations_blueprint.py` for various parameters of this task, including passing in custom annotation bucket definitions using the `annotation_buckets` YAML flag, being able to group multiple conversations into one HIT using the `subtasks_per_unit` flag, passing in onboarding data with answers, and being able to ask only for the final utterance as an annotation. + +**NOTE**: See [parlai/crowdsourcing/README.md](https://github.com/facebookresearch/ParlAI/blob/master/parlai/crowdsourcing/README.md) for general tips on running `parlai.crowdsourcing` tasks, such as how to specify your own YAML file of configuration settings, how to run tasks live, how to set parameters on the command line, etc.