Minor update to env.yaml

Author: andrewsong90

README.md

@@ -25,9 +25,9 @@ We introduce **M**ulti**M**odal **P**rototyping framework (**MMP**), where the r
 - 07/02/2024: The first version of MMP codebase is now live!
 
 ## Installation
-Please run the following command to create MMP conda environment.
+Once you clone the repo, please run the following command to create MMP conda environment.
 ```shell
-conda env create -f environment.yml
+conda env create -f env.yaml
 ```
 
 ## MMP Walkthrough

env.yaml

@@ -0,0 +1,78 @@
+name: mmp
+channels:
+  - defaults
+dependencies:
+  - _libgcc_mutex=0.1=main
+  - _openmp_mutex=5.1=1_gnu
+  - ca-certificates=2024.3.11=h06a4308_0
+  - ld_impl_linux-64=2.38=h1181459_1
+  - libffi=3.4.4=h6a678d5_1
+  - libgcc-ng=11.2.0=h1234567_1
+  - libgomp=11.2.0=h1234567_1
+  - libstdcxx-ng=11.2.0=h1234567_1
+  - ncurses=6.4=h6a678d5_0
+  - openssl=3.0.14=h5eee18b_0
+  - pip=24.0=py39h06a4308_0
+  - python=3.9.19=h955ad1f_1
+  - readline=8.2=h5eee18b_0
+  - setuptools=69.5.1=py39h06a4308_0
+  - sqlite=3.45.3=h5eee18b_0
+  - tk=8.6.14=h39e8969_0
+  - wheel=0.43.0=py39h06a4308_0
+  - xz=5.4.6=h5eee18b_1
+  - zlib=1.2.13=h5eee18b_1
+  - pip:
+    - certifi==2024.6.2
+    - charset-normalizer==3.3.2
+    - ecos==2.0.14
+    - einops==0.8.0
+    - faiss-gpu==1.7.2
+    - filelock==3.15.4
+    - fsspec==2024.6.1
+    - h5py==3.11.0
+    - huggingface-hub==0.23.4
+    - idna==3.7
+    - jinja2==3.1.4
+    - joblib==1.4.2
+    - markupsafe==2.1.5
+    - mpmath==1.3.0
+    - networkx==3.2.1
+    - numexpr==2.10.1
+    - numpy==1.26.4
+    - nvidia-cublas-cu12==12.1.3.1
+    - nvidia-cuda-cupti-cu12==12.1.105
+    - nvidia-cuda-nvrtc-cu12==12.1.105
+    - nvidia-cuda-runtime-cu12==12.1.105
+    - nvidia-cudnn-cu12==8.9.2.26
+    - nvidia-cufft-cu12==11.0.2.54
+    - nvidia-curand-cu12==10.3.2.106
+    - nvidia-cusolver-cu12==11.4.5.107
+    - nvidia-cusparse-cu12==12.1.0.106
+    - nvidia-nccl-cu12==2.20.5
+    - nvidia-nvjitlink-cu12==12.5.82
+    - nvidia-nvtx-cu12==12.1.105
+    - osqp==0.6.7.post0
+    - packaging==24.1
+    - pandas==2.2.2
+    - pot==0.9.4
+    - python-dateutil==2.9.0.post0
+    - pytz==2024.1
+    - pyyaml==6.0.1
+    - qdldl==0.1.7.post4
+    - regex==2024.5.15
+    - requests==2.32.3
+    - safetensors==0.4.3
+    - scikit-learn==1.5.0
+    - scikit-survival==0.23.0
+    - scipy==1.13.1
+    - six==1.16.0
+    - sympy==1.12.1
+    - threadpoolctl==3.5.0
+    - tokenizers==0.19.1
+    - torch==2.3.1
+    - tqdm==4.66.4
+    - transformers==4.42.3
+    - triton==2.3.1
+    - typing-extensions==4.12.2
+    - tzdata==2024.1
+    - urllib3==2.2.2

environment.yml

@@ -14,6 +14,7 @@ mode='faiss'  # 'faiss' or 'kmeans'
 n_proto=16  # Number of prototypes
 n_init=3  # Number of KMeans initializations to perform
 
+
 # Validity check for feat paths
 all_feat_dirs=""
 for dataroot_path in "${dataroots[@]}"; do

src/scripts/prototype/clustering.sh

@@ -6,7 +6,6 @@ config=$2
 ### Dataset Information
 declare -a dataroots=(
 	'path/to/tcga_brca'/
-
 )
 
 task='BRCA_survival'

src/scripts/survival/brca_surv.sh

@@ -108,16 +108,13 @@ def train(datasets, args):
         train_results = train_loop_survival(model, datasets['train'], optimizer, lr_scheduler, loss_fn,
                                             print_every=args.print_every, accum_steps=args.accum_steps)
 
-        writer = log_dict_tensorboard(writer, train_results, 'train/', epoch)
 
         ### Validation Loop (Optional)
         if 'val' in datasets.keys():
             print('#' * 11, f'VAL Epoch: {epoch}', '#' * 11)
             val_results, _ = validate_survival(model, datasets['val'], loss_fn,
                                                    print_every=args.print_every, verbose=True)
 
-            writer = log_dict_tensorboard(writer, val_results, 'val/', epoch)
-
             ### Check Early Stopping (Optional)
             if early_stopper is not None:
                 if args.es_metric == 'loss':

src/training/trainer.py

@@ -80,7 +80,8 @@ def cluster(data_loader, n_proto, n_iter, n_init=5, feature_dim=1024, n_proto_pa
                               verbose=True, 
                               max_points_per_centroid=n_proto_patches,
                               gpu=numOfGPUs)
-        kmeans.train(patches)
+        
+        kmeans.train(patches.numpy())
         weight = kmeans.centroids[np.newaxis, ...]
 
     else: