update

Dockerfile

@@ -12,7 +12,11 @@ RUN pip install --no-cache-dir --upgrade -r /code/requirements.txt
 
 RUN mkdir -p /.cache/huggingface/hub && \
     chmod a+rw /.cache/huggingface/hub && \
-    echo made cache dir read-write
+    echo made huggingface cache dir read-write
+
+RUN mkdir -p /.cache/gorilla && \
+    chmod a+rw /.cache/gorilla && \
+    echo made gorilla cache dir read-write
 
 # Set up a new user named "user" with user ID 1000
 RUN useradd -m -u 1000 user

main.py

@@ -29,7 +29,7 @@ def get_prompt(user_query: str, functions: list = []) -> str:
 device : str = "cuda:0" if torch.cuda.is_available() else "cpu"
 torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32
 
-result = subprocess.run('ls -lH', shell=True, capture_output=True, text=True)
+result = subprocess.run('pwd && ls -lH && find /.cache/huggingface/hub && find /.cache/gorilla', shell=True, capture_output=True, text=True)
 log(result.stdout)
 
 model_id : str = "gorilla-llm/gorilla-openfunctions-v1"
@@ -37,11 +37,13 @@ log('AutoTokenizer.from_pretrained ...')
 tokenizer = AutoTokenizer.from_pretrained(model_id)
 log('AutoModelForCausalLM.from_pretrained ...')
 model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype=torch_dtype, low_cpu_mem_usage=True)
+log('AutoModelForCausalLM.save_pretrained ...')
+model.save_pretrained('/.cache/gorilla')
 
-result = subprocess.run('ls -lH', shell=True, capture_output=True, text=True)
+result = subprocess.run('pwd && ls -lH && find /.cache/huggingface/hub && find /.cache/gorilla', shell=True, capture_output=True, text=True)
 log(result.stdout)
 
-log('mode.to(device) ...')
+log('model.to(device) ...')
 model.to(device)
 
 log('FastAPI setup ...')