adapted Dockerfile and yaml file for Mamba

Dockerfile

@@ -1,5 +1,5 @@
-# Use a Conda-enabled base image
-FROM continuumio/miniconda3:latest
+# Use a Mambaforge base image for faster and more memory-efficient solving
+FROM condaforge/mambaforge:latest
 
 # Set the working directory in the container
 WORKDIR /app
@@ -7,14 +7,15 @@ WORKDIR /app
 # Copy the environment.yml file into the container
 COPY environment.yml .
 
-# Create the Conda environment from environment.yml
-# --name research-agent-gpu : Use the name defined in your environment.yml
-# -f environment.yml       : Use the copied file
-# Conda cleans up its cache to reduce image size
-RUN conda env create -f environment.yml && conda clean --all
+# Create the Conda environment using mamba and clean up
+# mamba env create is the equivalent of conda env create but faster and less memory intensive
+# The --yes flag answers yes to prompts
+# mamba clean --all clears caches
+RUN mamba env create --yes -f environment.yml && mamba clean --all
 
 # Activate the Conda environment and set it as the default for subsequent commands
-ENV PATH=/opt/conda/envs/research-agent-gpu/bin:$PATH
+# Ensure the environment name matches what's in your environment.yml
+ENV PATH /opt/conda/envs/research-agent-gpu/bin:$PATH
 
 # Copy the rest of your application code
 COPY . .
@@ -23,5 +24,4 @@ COPY . .
 EXPOSE 7860
 
 # Command to run your application when the container starts
-# Make sure app.py is in the root of your /app directory inside the container
 CMD ["python", "app.py"]

environment.yml

@@ -6,8 +6,8 @@
 
 name: research-agent-gpu
 channels:
-  - pytorch
-  - nvidia
+  # - pytorch
+  # - nvidia
   - conda-forge
   - defaults
 dependencies: