yujunjun/haystack
1
0
Fork 0
mirror of https://github.com/deepset-ai/haystack.git synced 2025-10-14 17:38:24 +00:00
Code Issues Packages Projects Releases Wiki Activity

add img to readme. update FARM api format

Browse Source
This commit is contained in:
Malte Pietsch 2019-11-25 14:39:36 +01:00
parent 45365e6b99
commit 6f8a015145
5 changed files with 13 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
README.rst
View File

@ -12,11 +12,11 @@ Haystack is designed in a modular way and is tightly integrated with the `FARM <
Swap your models easily from BERT to roBERTa and scale the database from dev (Sqlite) to production (PostgreSQL, elasticsearch ...). Swap your models easily from BERT to roBERTa and scale the database from dev (Sqlite) to production (PostgreSQL, elasticsearch ...).
Core Features Core Features
========== =============
- **Most powerful models**: Utilize all the latest transformer based models (BERT, ALBERT roBERTa ...) - **Most powerful models**: Utilize all the latest transformer based models (BERT, ALBERT roBERTa ...)
- **Modular & future-proof**: Avoid technical debt. With haystack you can easily switch to newer models once they get published. - **Modular & future-proof**: Avoid technical debt. With haystack you can easily switch to newer models once they get published.
- **Developer friendly**: Easy to debug, extend and modify - **Developer friendly**: Easy to debug, extend and modify.
- **Scalable**: Switch from dev to production within minutes. - **Scalable**: Switch from dev to production within minutes.
Components Components
========== ==========
@ -37,9 +37,14 @@ Installation
------------ ------------
There are two ways to install: There are two ways to install:
* (recommended) from source, :code:`git clone <url>` and run :code:`pip install [--editable] .` from the root of the repositry. * (recommended) from source, :code:`git clone <url>` and run :code:`pip install [--editable] .` from the root of the repositry.
* from PyPI, do a :code:`pip install farm_haystack` * from PyPI, do a :code:`pip install farm_haystack`
Usage
-----
.. image:: https://raw.githubusercontent.com/deepset-ai/haystack/master/docs/img/code_snippet_usage.png
Configuration Configuration
------------- -------------
The configuration can be supplied in a :code:`qa_config.py` placed in the PYTHONPATH. Alternatively, the :code:`DATABASE_URL` can also be set an an environment variable. The configuration can be supplied in a :code:`qa_config.py` placed in the PYTHONPATH. Alternatively, the :code:`DATABASE_URL` can also be set an an environment variable.

BIN
docs/img/code_snippet_usage.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 229 KiB

0
example.py Normal file
View File

2
farm_haystack/__init__.py
View File

@ -60,7 +60,7 @@ class Finder:
retrieved_scores, question, candidate_doc_ids=candidate_doc_ids retrieved_scores, question, candidate_doc_ids=candidate_doc_ids
) )
results = self.reader.predict(inference_dicts, top_k=top_k_reader) results = self.reader.predict(inference_dicts, top_k=top_k_reader)
return results return results["results"]
def _convert_retrieved_text_to_reader_format( def _convert_retrieved_text_to_reader_format(
self, retrieved_scores, question, candidate_doc_ids=None, verbose=True self, retrieved_scores, question, candidate_doc_ids=None, verbose=True

3
farm_haystack/utils.py
View File

@ -18,7 +18,8 @@ def create_db():
db.session.create_all() db.session.create_all()
def print_answers(answers, details="all"): def print_answers(results, details="all"):
answers = results[0]["answers"]
pp = pprint.PrettyPrinter(indent=4) pp = pprint.PrettyPrinter(indent=4)
if details != "all": if details != "all":
if details == "minimal": if details == "minimal":