Loreto Parisi

So what I can see from the logs it's a sequence of bounces between `Connecting` and `Disconnected` states ``` Disconnected Connecting Disconnected Connecting Disconnected Connecting Disconnected Connecting Disconnected Connecting Disconnected...

@bittlingmayer nice idea! go on, I will push the repo there!

@bittlingmayer great, I will check it out next week, in the meanwhile I have joined the org. Next step I will push the repo in `js/` folder.

@bittlingmayer fixed the team member. Give me few days I will complete the move very busy week this one!

@bittlingmayer no worries, I know that one :) going to do this this week

@Uzay-G correct I will provide a modified WASM module for browser based inference.

@ebuildy yes it is, Basically it's a BPE tokenizer, so it turns sentences into codes, and map the codes to words into a vocabulary (actually not words, more like subwords)....

In my case I put it here ```python tf.reset_default_graph() if not sess: sess = gpt2.start_tf_sess() else: sess = gpt2.reset_session(sess) gpt2.load_gpt2(sess, run_name=run_name) ``` and it perfectly worked! Thanks!

When in Python3 on macOS I will get a crash of multiprocessing lib ``` Process: Python [67875] Path: /usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/Resources/Python.app/Contents/MacOS/Python Identifier: Python Version: 3.7.3 (3.7.3) Code Type: X86-64 (Native) Parent Process:...

Hey guys I tried to reproduce `to_serialized` and `from_serialized` ```python # doc.to_serialized json_str = json.dumps(doc.to_dict(), indent=4, ensure_ascii = True) serialized_string = pickle.dumps((sequence, json_str)) # Document.from_serialized (text, sentences) = pickle.loads(serialized_string) print(text,...