use fastchat conversations template (#578)

* use fastchat conversations template

* require fastchat (fschat) pip install

* handle roles dynamically from conversation

* tweak fastchat conversation with a monkeypatch to get individual turns

* fix up so it works with multiple conversation styles, and don't strip the turns

* fix sharegpt fixture now that we're using a more correct tokenization

* use a new prompter and support fastchat conversation type

* use sharegpt from prompt strategies now

* update docs, add chatml template

* add a newline after im_end token

* ensure we correctly set system message

* update per PR feedback to handle deprecated sharegpt types

* don't add duplicate wandb req

* make sharegpt fields configurable from yml

* llama2 fixes

* don't fail fatally when turns are improper
This commit is contained in:
Wing Lian
2023-09-27 12:10:45 -04:00
committed by GitHub
parent 60c7c48c97
commit e7d3e2dbb6
13 changed files with 324 additions and 112 deletions

View File

@@ -180,7 +180,7 @@ Have dataset(s) in one of the following format (JSONL recommended):
```json
{"instruction": "...", "input": "...", "output": "..."}
```
- `sharegpt:chat`: conversations where `from` is `human`/`gpt`
- `sharegpt`: conversations where `from` is `human`/`gpt`
```json
{"conversations": [{"from": "...", "value": "..."}]}
```
@@ -269,11 +269,11 @@ Have dataset(s) in one of the following format (JSONL recommended):
```json
{"prompt": "...", "generation": "..."}
```
- `sharegpt_simple.load_role`: conversations where `role` is used instead of `from`
- `sharegpt.load_role`: conversations where `role` is used instead of `from`
```json
{"conversations": [{"role": "...", "value": "..."}]}
```
- `sharegpt_simple.load_guanaco`: conversations where `from` is `prompter`/`assistant` instead of default sharegpt
- `sharegpt.load_guanaco`: conversations where `from` is `prompter`/`assistant` instead of default sharegpt
```json
{"conversations": [{"from": "...", "value": "..."}]}
```
@@ -443,6 +443,7 @@ datasets:
data_files: # Optional[str] path to source data files
shards: # Optional[int] number of shards to split data into
name: # Optional[str] name of dataset configuration to load
conversation: # Optional[str] fastchat conversation type, only used with type: sharegpt
# custom user prompt
- path: repo