Yuvraj-Dhepe
commented
2 months ago cmd_line_opt.txt This is the cmd opt with debug option
Closed Yuvraj-Dhepe closed 1 month ago
Yuvraj-Dhepe
commented
2 months ago cmd_line_opt.txt This is the cmd opt with debug option
jhgoebbert
commented
2 months ago This is an issue with your Dask/distributed version. Checking the backtrace
File "~/micromamba/envs/audio_dl/lib/python3.9/site-packages/distributed/worker_memory.py", line 56, in <module>
WorkerDataParameter: TypeAlias = Union[and the code https://github.com/dask/distributed/blob/main/distributed/worker_memory.py#L56 you can find the issue here: https://github.com/dask/distributed/issues/8349
Upgrading to Python 3.9.2 seems to be the solution.
Yuvraj-Dhepe
commented
2 months ago Perfect Thanks @jhgoebbert It worked out.
Yuvraj-Dhepe
commented
2 months ago I was able to use the chat option, however the leanr option seems to throw errors.
/learn chat handler resolved in 18 ms.
2024-07-27 22:13:28,026 - distributed.worker - WARNING - Compute Failed
Key: embed_chunk-b36c3fe5-b881-4826-9ca9-d4ba836341a3
State: executing
Function: execute_task
args: ((<function embed_chunk at 0x7f1439475ca0>, Document(metadata={'path': '~/work/Projects/Audio_DL/dcase2023-audio-retrieval/utils/data_utils.py', 'sha256': b'\xa9!\x9d\x88e\xef\x966R\xbc\xb0\x8d\x9bf\xfb$\x1am\xa9\xd3\xc1\xf8PJ\xa1\x84-\x82\xc7N\x10\x87', 'extension': '.py'}, page_content='import os\nimport pickle\nfrom ast import literal_eval\n\nimport h5py\nimport nltk\nimport numpy as np\nimport pandas as pd\nimport torch\nimport torch.nn.functional as F\nfrom torch.utils.data import Dataset\n\nstopwords = nltk.corpus.stopwords.words("english")\n\n\nclass Vocabulary(object):\n\n def __init__(self):\n self.key2vec = {}\n self.key2id = {}\n self.id = 0\n\n def add_key(self, key, key_vector):\n if key not in self.key2id:\n self.key2vec[key] = key_vector\n self.key2id[key] = self.id\n self.id += 1\n\n def __call__(self, key):\n return self.key2id[key]\n\n def __len__(self):\n return len(self
kwargs: {}
Exception: 'ValueError("Error raised by inference endpoint: HTTPConnectionPool(host=\'localhost\', port=11434): Max retries exceeded with url: /api/embeddings (Caused by NewConnectionError(\'<urllib3.connection.HTTPConnection object at 0x7f142a5f1b50>: Failed to establish a new connection: [Errno 111] Connection refused\'))")'
Traceback: ' File "~/micromamba/envs/audio_dl/lib/python3.9/site-packages/jupyter_ai/document_loaders/directory.py", line 165, in embed_chunk\n embedding = em.embed_query(content)\n File "~/micromamba/envs/audio_dl/lib/python3.9/site-packages/langchain_community/embeddings/ollama.py", line 224, in embed_query\n embedding = self._embed([instruction_pair])[0]\n File "~/micromamba/envs/audio_dl/lib/python3.9/site-packages/langchain_community/embeddings/ollama.py", line 199, in _embed\n return [self._process_emb_response(prompt) for prompt in iter_]\n File "~/micromamba/envs/audio_dl/lib/python3.9/site-packages/langchain_community/embeddings/ollama.py", line 199, in <listcomp>\n return [self._process_emb_response(prompt) for prompt in iter_]\n File "~/micromamba/envs/audio_dl/lib/python3.9/site-packages/langchain_community/embeddings/ollama.py", line 170, in _process_emb_response\n raise ValueError(f"Error raised by inference endpoint: {e}")\n'
2024-07-27 22:13:28,026 - distributed.worker - WARNING - Compute Failed
Key: embed_chunk-c20dc18c-5eac-4fc8-b0c8-8bc59710e107
State: executing
Function: execute_task
args: ((<function embed_chunk at 0x7f1439475ca0>, Document(metadata={'path': '~/work/Projects/Audio_DL/dcase2023-audio-retrieval/utils/data_utils.py', 'sha256': b'\xa9!\x9d\x88e\xef\x966R\xbc\xb0\x8d\x9bf\xfb$\x1am\xa9\xd3\xc1\xf8PJ\xa1\x84-\x82\xc7N\x10\x87', 'extension': '.py'}, page_content='class AudioTextDataset(Dataset):\n\n def __init__(self, **kwargs):\n self.audio_data = kwargs["audio_data"]\n self.text_data = kwargs["text_data"]\n self.text_vocab = kwargs["text_vocab"]\n self.text_level = kwargs["text_level"]\n\n def __getitem__(self, index):\n item = self.text_data.iloc[index]\n\n audio_vec = torch.as_tensor(self.audio_data[item["fid"]][()])\n\n text_vec = None\n\n if self.text_level == "word":\n text_vec = torch.as_tensor([self.text_vocab(key) for key in item["tokens"] if key not in stopwords])\n\n elif self.text_level == "sentence":\n text_vec = torch.as_tensor([self.text_vocab(
kwargs: {}
Exception: 'ValueError("Error raised by inference endpoint: HTTPConnectionPool(host=\'localhost\', port=11434): Max retries exceeded with url: /api/embeddings (Caused by NewConnectionError(\'<urllib3.connection.HTTPConnection object at 0x7f142a5f3670>: Failed to establish a new connection: [Errno 111] Connection refused\'))")'
Traceback: ' File "~/micromamba/envs/audio_dl/lib/python3.9/site-packages/jupyter_ai/document_loaders/directory.py", line 165, in embed_chunk\n embedding = em.embed_query(content)\n File "~/micromamba/envs/audio_dl/lib/python3.9/site-packages/langchain_community/embeddings/ollama.py", line 224, in embed_query\n embedding = self._embed([instruction_pair])[0]\n File "~/micromamba/envs/audio_dl/lib/python3.9/site-packages/langchain_community/embeddings/ollama.py", line 199, in _embed\n return [self._process_emb_response(prompt) for prompt in iter_]\n File "~/micromamba/envs/audio_dl/lib/python3.9/site-packages/langchain_community/embeddings/ollama.py", line 199, in <listcomp>\n return [self._process_emb_response(prompt) for prompt in iter_]\n File "~/micromamba/envs/audio_dl/lib/python3.9/site-packages/langchain_community/embeddings/ollama.py", line 170, in _process_emb_response\n raise ValueError(f"Error raised by inference endpoint: {e}")\n'
This is my jupyter ai settings, I also do have the embeddings model running at the provided host and port.
jhgoebbert
commented
2 months ago If you check the stacktrace carefully you can see that this is more an issue belonging to langchain / docker. A quick look there will lead you for example to this: https://github.com/langchain-ai/langchain/issues/19074 """The issue is with the docker compose configuration."""
sdhdez
commented
1 month ago Hi,
I'm encountering a similar issue with Ollama running behind a reverse proxy. Interestingly, I haven't configured port 11434 at all and I'm still having this error:
ConnectionError: HTTPConnectionPool(host='localhost', port=11434): Max retries exceeded with url: /api/generate (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x7fac48882d80>: Failed to establish a new connection: [Errno 111] Connection refused'))It seems that jupyter-ai is ignoring the base_url specified in the interface settings. Instead, it's using the hardcoded base_url: str = "http://localhost:11434" :
~$ find . -type f -name "*.py" | xargs grep "localhost:1111434" --color
./.conda/pkgs/langchain-community-0.2.12-pyhd8ed1ab_0/site-packages/langchain_community/llms/ollama.py: base_url: str = "http://localhost:11434"
./.conda/pkgs/langchain-community-0.2.12-pyhd8ed1ab_0/site-packages/langchain_community/embeddings/ollama.py: base_url: str = "http://localhost:11434"
./.conda/envs/MyPyEnv/lib/python3.12/site-packages/langchain_community/llms/ollama.py: base_url: str = "http://localhost:11434"
./.conda/envs/MyPyEnv/lib/python3.12/site-packages/langchain_community/embeddings/ollama.py: base_url: str = "http://localhost:11434"Update: It seems the issue is addressed here #902.
srdas
commented
1 month ago Issue #902 has raised this issue and it is fixed in PR #904 - closing this issue.
Description
I installed the jupyter ai extension in a mamba environment using
pip install jupyter-ai, I got jupyter lab installed automatically with this command. However, when I loaded the jupyter lab, I getIn the terminal I get the following errors when I start the jupyter lab
jupyter lab --no-browserReproduce
1) Create a simple mamba environment: mamba create -f environment.yaml
4) Start the jupyter lab server:
jupyter lab --no-browser5) Later open the chat widget to see if the chat works
Expected behavior
Context
Linux <user-name> 5.15.150.1-microsoft-standard-WSL2+ #1 SMP Sun Apr 7 22:57:26 CEST 2024 x86_64 x86_64 x86_64 GNU/LinuxTroubleshoot Output
$PATH: /home//micromamba/envs/audio_dl/bin
/home//.local/bin
/home//.bun/bin
/home/linuxbrew/.linuxbrew/bin
/home/linuxbrew/.linuxbrew/sbin
/home//micromamba/condabin
/usr/local/cuda-12.4/bin
/usr/local/sbin
/usr/local/bin
/usr/sbin
/usr/bin
/sbin
/bin
/usr/games
/usr/local/games
/usr/lib/wsl/lib
/mnt/c/Program Files/Microsoft MPI/Bin/
/mnt/c/WINDOWS/system32
/mnt/c/WINDOWS
/mnt/c/WINDOWS/System32/Wbem
/mnt/c/WINDOWS/System32/WindowsPowerShell/v1.0/
/mnt/c/WINDOWS/System32/OpenSSH/
/mnt/c/Program Files (x86)/NVIDIA Corporation/PhysX/Common
/mnt/c/Program Files/NVIDIA Corporation/NVIDIA App/NvDLISR
/mnt/c/Program Files/usbipd-win/
/mnt/d/WSL/work/Git/cmd
/mnt/c/Program Files/PowerShell/7/
/mnt/c/Users//AppData/Local/Microsoft/WindowsApps
/mnt/c/Users//AppData/Local/Programs/Microsoft VS Code/bin
/mnt/c/Users//AppData/Local/Programs/oh-my-posh/bin
/mnt/c/Users//AppData/Local/GitHubDesktop/bin
/mnt/c/Users//AppData/Local/Programs/Ollama
/mnt/d/WSL/work/micromamba
/snap/bin
sys.path: /home//micromamba/envs/audio_dl/bin
/home//micromamba/envs/audio_dl/lib/python39.zip
/home//micromamba/envs/audio_dl/lib/python3.9
/home//micromamba/envs/audio_dl/lib/python3.9/lib-dynload
/home//micromamba/envs/audio_dl/lib/python3.9/site-packages
sys.executable: /home//micromamba/envs/audio_dl/bin/python3.9
sys.version: 3.9.0 | packaged by conda-forge | (default, Nov 26 2020, 07:57:39) [GCC 9.3.0]
platform.platform(): Linux-5.15.150.1-microsoft-standard-WSL2+-x86_64-with-glibc2.35
which -a jupyter: /home//micromamba/envs/audio_dl/bin/jupyter
pip list: Package Version
Command Line Output
Browser Output