docker changes

Dockerfile

@@ -23,17 +23,21 @@ RUN conda install python~=3.10.12 pip && \
 
 VOLUME /cache
 ENV PETALS_CACHE=/cache
-#COPY pip.freeze petals/pip.freeze
-#RUN pip install --no-cache-dir -r petals/pip.freeze
-#COPY pip.freeze2 petals/pip.freeze2
-#RUN pip install --no-cache-dir -r petals/pip.freeze2
 
+ADD pip.freeze petals/pip.freeze
+RUN pip install --no-cache-dir -r petals/pip.freeze
+ADD pip2.freeze petals/pip2.freeze
+RUN pip install --no-cache-dir -r petals/pip2.freeze
 
 ADD tests petals/tests
-ADD LICENSE  README.md pyproject.toml setup.cfg petals/
 ADD src petals/src
+ADD LICENSE  README.md pyproject.toml setup.cfg petals/
+
 RUN pip install --no-cache-dir -e petals
 #RUN pip install --no-cache-dir --upgrade transformers==4.34.0
 
 WORKDIR /home/petals/
+
+RUN pip freeze > pip.freeze.new
+
 CMD python -m petals.cli.run_server --port 31331  --num_blocks=1 Maykeye/TinyLLama-v0 

docker-compose.yml

@@ -8,7 +8,6 @@ services:
     network_mode: host
     ipc: host
     restart: unless-stopped
-    env_file: ./.env
     
   health:
     profiles: ["core"]
@@ -19,7 +18,6 @@ services:
     ports:
       - "8009:5000"
     command: flask run --host=0.0.0.0 --port=5000
-    env_file: ./.env
 
   tinyllama_local_gpu:
     profiles: ["local","gpu"]
@@ -38,9 +36,30 @@ services:
               count: 1
               capabilities: [gpu]
     restart: always
-    env_file: ./.env
 
 #
+  envmodel_local_gpu:
+    profiles: ["local","gpu"]
+    build: .
+    environment:
+      - MODEL=${MODEL}
+    depends_on:
+       - backbone
+       # xai-org/grok-1
+       # hpcai-tech/grok-1
+       # keyfan/grok-1-hf
+    command: python -m petals.cli.run_server --port 31331  --num_blocks=1  $MODEL  --initial_peers $INITIAL_PEERS  --device=$DEVICE
+    ports:
+      - "31331:31331"
+    restart: always
+    deploy:
+      resources:
+        reservations:
+          devices:
+            - driver: nvidia
+              count: 1
+              capabilities: [gpu]
+>>>>>>> 79bb253 (docker changes)
   tinymixtral_local_gpu:
     profiles: ["local","gpu"]
     build: .
@@ -50,7 +69,7 @@ services:
     ports:
       - "31331:31331"
     restart: always
-    env_file: ./.env
+
     deploy:
       resources:
         reservations:
@@ -68,7 +87,7 @@ services:
     ports:
       - "31331:31331"
     restart: always
-    env_file: ./.env
+
     
   tinyllamacpu:
     profiles: ["tinyllama","cpu"]
@@ -79,7 +98,7 @@ services:
     ports:
       - "31331:31331"
     restart: always
-    env_file: ./.env
+
     
   tinyllamagpu:
     profiles: ["core"]
@@ -97,7 +116,7 @@ services:
               count: 1
               capabilities: [gpu]
     restart: always
-    env_file: ./.env
+
     
   tinyllamatpu:
     profiles: ["tpu"]
@@ -108,7 +127,7 @@ services:
       - "31333:31333"
     command: python -m petals.cli.run_server --port 31333  --num_blocks=1 Maykeye/TinyLLama-v0 --initial_peers $INITIAL_PEERS  --device=$DEVICE
     restart: always
-    env_file: ./.env
+
 
   debug_health:
     profiles: ["debug"]
@@ -127,7 +146,7 @@ services:
 #    image: petals-inference-test
     ports:
       - "8010:5000"
-    env_file: ./.env
+
     command: gunicorn app:app --bind 0.0.0.0:5000 --worker-class gthread --threads 100 --timeout 1000
 
 

pip.freeze

@@ -8,16 +8,17 @@ charset-normalizer==3.3.2
 ConfigArgParse==1.7
 cpufeature==0.2.1
 cryptography==42.0.5
+
 Dijkstar==2.6.0
 filelock==3.13.4
 fsspec==2024.3.1
 grpcio==1.62.1
 grpcio-tools==1.62.1
-hivemind==1.1.10.post2
-#huggingface-hub==0.22.2
 humanfriendly==10.0
 idna==3.7
 Jinja2==3.1.3
+jsonpointer==2.1
+
 MarkupSafe==2.1.5
 mpmath==1.3.0
 msgpack==1.0.8
@@ -37,6 +38,7 @@ nvidia-cusparse-cu12==12.1.0.106
 nvidia-nccl-cu12==2.19.3
 nvidia-nvjitlink-cu12==12.4.127
 nvidia-nvtx-cu12==12.1.105
+
 packaging==24.0
 peft==0.5.0
 prefetch-generator==1.0.3
@@ -48,6 +50,7 @@ pymultihash==0.8.2
 PyYAML==6.0.1
 regex==2023.12.25
 requests==2.31.0
+
 safetensors==0.4.3
 scipy==1.13.0
 sentencepiece==0.2.0
@@ -57,14 +60,14 @@ speedtest-cli==2.1.3
 sympy==1.12
 tensor-parallel==1.0.23
 
-torch==2.2.2
 tqdm==4.66.2
-#transformers==4.38.2
-
 triton==2.2.0
 typing_extensions==4.11.0
 urllib3==2.2.1
 uvloop==0.19.0
 varint==1.0.2
-
+triton==2.2.0
+typing_extensions==4.11.0
+uvloop==0.19.0
+varint==1.0.2
 

pip2.freeze

@@ -0,0 +1,6 @@
+transformers==4.38.2
+huggingface-hub==0.22.2
+hivemind==1.1.10.post2
+peft==0.5.0
+tokenizers==0.15.2
+torch==2.2.2