update links

mit-han-lab · Jul 19, 2023 · cf2ff3a · cf2ff3a
1 parent 101fa22
commit cf2ff3a
Show file tree

Hide file tree

Showing 10 changed files with 43 additions and 43 deletions.
diff --git a/README.md b/README.md
@@ -41,7 +41,7 @@
 ![](figures/diverse_hardware.png)
 
 ## OFA-ResNet50 [[How to use]](https://github.com/mit-han-lab/once-for-all/blob/master/tutorial/ofa_resnet50_example.ipynb)
-<img src="https://hanlab.mit.edu/files/OnceForAll/figures/ofa_resnst50_results.png" width="60%" />
+<img src="figures/ofa_resnst50_results.png" width="60%" />
 
 ## How to use / evaluate **OFA Networks**
 ### Use

diff --git a/figures/ofa_resnst50_results.png b/figures/ofa_resnst50_results.png
diff --git a/ofa/model_zoo.py b/ofa/model_zoo.py
@@ -23,8 +23,9 @@
 ]
 
 
-def ofa_specialized(net_id, pretrained=True):
-    url_base = "https://hanlab.mit.edu/files/OnceForAll/ofa_specialized/"
+def ofa_specialized(net_id: str, pretrained=True):
+    url_base = "https://raw.githubusercontent.com/han-cai/files/master/ofa/ofa_specialized/"
+    net_id = net_id.replace("@", "-")
     net_config = json.load(
         open(
             download_url(
@@ -88,12 +89,12 @@ def ofa_net(net_id, pretrained=True):
             expand_ratio_list=[0.2, 0.25, 0.35],
             width_mult_list=[0.65, 0.8, 1.0],
         )
-        net_id = "ofa_resnet50_d=0+1+2_e=0.2+0.25+0.35_w=0.65+0.8+1.0"
+        net_id = "ofa_resnet50_d0+1+2_e0.2+0.25+0.35_w0.65+0.8+1.0"
     else:
         raise ValueError("Not supported: %s" % net_id)
 
     if pretrained:
-        url_base = "https://hanlab.mit.edu/files/OnceForAll/ofa_nets/"
+        url_base = "https://raw.githubusercontent.com/han-cai/files/master/ofa/ofa_nets/"
         init = torch.load(
             download_url(url_base + net_id, model_dir=".torch/ofa_nets"),
             map_location="cpu",
@@ -104,13 +105,13 @@ def ofa_net(net_id, pretrained=True):
 
 def proxylessnas_net(net_id, pretrained=True):
     net = proxyless_base(
-        net_config="https://hanlab.mit.edu/files/proxylessNAS/%s.config" % net_id,
+        net_config="https://raw.githubusercontent.com/han-cai/files/master/proxylessnas/%s.config" % net_id,
     )
     if pretrained:
         net.load_state_dict(
             torch.load(
                 download_url(
-                    "https://hanlab.mit.edu/files/proxylessNAS/%s.pth" % net_id
+                    "https://raw.githubusercontent.com/han-cai/files/master/proxylessnas/%s.pth" % net_id
                 ),
                 map_location="cpu",
             )["state_dict"]

diff --git a/ofa/nas/efficiency_predictor/latency_lookup_table.py b/ofa/nas/efficiency_predictor/latency_lookup_table.py
@@ -25,7 +25,7 @@ class LatencyTable(object):
     def __init__(
         self,
         local_dir="~/.ofa/latency_tools/",
-        url="https://hanlab.mit.edu/files/proxylessNAS/LatencyTools/mobile_trim.yaml",
+        url="https://raw.githubusercontent.com/han-cai/files/master/proxylessnas/mobile_trim.yaml",
     ):
         if url.startswith("http"):
             fname = download_url(url, local_dir, overwrite=True)

diff --git a/ofa/tutorial/accuracy_predictor.py b/ofa/tutorial/accuracy_predictor.py
@@ -37,7 +37,7 @@ def __init__(self, pretrained=True, device="cuda:0"):
         if pretrained:
             # load pretrained model
             fname = download_url(
-                "https://hanlab.mit.edu/files/OnceForAll/tutorial/acc_predictor.pth"
+                "https://raw.githubusercontent.com/han-cai/files/master/ofa/acc_predictor.pth"
             )
             self.model.load_state_dict(
                 torch.load(fname, map_location=torch.device("cpu"))

diff --git a/ofa/tutorial/latency_table.py b/ofa/tutorial/latency_table.py
@@ -6,7 +6,7 @@ class LatencyEstimator(object):
     def __init__(
         self,
         local_dir="~/.hancai/latency_tools/",
-        url="https://hanlab.mit.edu/files/proxylessNAS/LatencyTools/mobile_trim.yaml",
+        url="https://raw.githubusercontent.com/han-cai/files/master/proxylessnas/mobile_trim.yaml",
     ):
         if url.startswith("http"):
             fname = download_url(url, local_dir, overwrite=True)
@@ -198,8 +198,7 @@ def __init__(self, device="note10", resolutions=(160, 176, 192, 208, 224)):
 
         for image_size in resolutions:
             self.latency_tables[image_size] = LatencyEstimator(
-                url="https://hanlab.mit.edu/files/OnceForAll/tutorial/latency_table@%s/%d_lookup_table.yaml"
-                % (device, image_size)
+                url=f"https://raw.githubusercontent.com/han-cai/files/master/ofa/{device}/{image_size}_lookup_table.yaml"
             )
             print("Built latency table for image size: %d." % image_size)
 

diff --git a/setup.py b/setup.py
@@ -43,7 +43,7 @@
     name="ofa",
     version=VERSION,
     author="MTI HAN LAB ",
-    author_email="hanlab.eecs+github@gmail.com",
+    author_email="hcai.hm@gmail.com",
     url="https://github.com/mit-han-lab/once-for-all",
     description="Once for All: Train One Network and Specialize it for Efficient Deployment.",
     long_description=readme,

diff --git a/train_ofa_net.py b/train_ofa_net.py
@@ -137,7 +137,7 @@
     torch.cuda.set_device(hvd.local_rank())
 
     args.teacher_path = download_url(
-        "https://hanlab.mit.edu/files/OnceForAll/ofa_checkpoints/ofa_D4_E6_K7",
+        "https://raw.githubusercontent.com/han-cai/files/master/ofa/ofa_checkpoints/ofa_D4_E6_K7",
         model_dir=".torch/ofa_checkpoints/%d" % hvd.rank(),
     )
 
@@ -255,7 +255,7 @@
         validate_func_dict["ks_list"] = sorted(args.ks_list)
         if distributed_run_manager.start_epoch == 0:
             args.ofa_checkpoint_path = download_url(
-                "https://hanlab.mit.edu/files/OnceForAll/ofa_checkpoints/ofa_D4_E6_K7",
+                "https://raw.githubusercontent.com/han-cai/files/master/ofa/ofa_checkpoints/ofa_D4_E6_K7",
                 model_dir=".torch/ofa_checkpoints/%d" % hvd.rank(),
             )
             load_models(
@@ -284,12 +284,12 @@
 
         if args.phase == 1:
             args.ofa_checkpoint_path = download_url(
-                "https://hanlab.mit.edu/files/OnceForAll/ofa_checkpoints/ofa_D4_E6_K357",
+                "https://raw.githubusercontent.com/han-cai/files/master/ofa/ofa_checkpoints/ofa_D4_E6_K357",
                 model_dir=".torch/ofa_checkpoints/%d" % hvd.rank(),
             )
         else:
             args.ofa_checkpoint_path = download_url(
-                "https://hanlab.mit.edu/files/OnceForAll/ofa_checkpoints/ofa_D34_E6_K357",
+                "https://raw.githubusercontent.com/han-cai/files/master/ofa/ofa_checkpoints/ofa_D34_E6_K357",
                 model_dir=".torch/ofa_checkpoints/%d" % hvd.rank(),
             )
         train_elastic_depth(train, distributed_run_manager, args, validate_func_dict)
@@ -300,12 +300,12 @@
 
         if args.phase == 1:
             args.ofa_checkpoint_path = download_url(
-                "https://hanlab.mit.edu/files/OnceForAll/ofa_checkpoints/ofa_D234_E6_K357",
+                "https://raw.githubusercontent.com/han-cai/files/master/ofa/ofa_checkpoints/ofa_D234_E6_K357",
                 model_dir=".torch/ofa_checkpoints/%d" % hvd.rank(),
             )
         else:
             args.ofa_checkpoint_path = download_url(
-                "https://hanlab.mit.edu/files/OnceForAll/ofa_checkpoints/ofa_D234_E46_K357",
+                "https://raw.githubusercontent.com/han-cai/files/master/ofa/ofa_checkpoints/ofa_D234_E46_K357",
                 model_dir=".torch/ofa_checkpoints/%d" % hvd.rank(),
             )
         train_elastic_expand(train, distributed_run_manager, args, validate_func_dict)

diff --git a/tutorial/ofa.ipynb b/tutorial/ofa.ipynb
diff --git a/tutorial/ofa_resnet50_example.ipynb b/tutorial/ofa_resnet50_example.ipynb
@@ -16,6 +16,12 @@
   {
    "cell_type": "code",
    "execution_count": 5,
+   "metadata": {
+    "collapsed": false,
+    "pycharm": {
+     "name": "#%%\n"
+    }
+   },
    "outputs": [
     {
      "name": "stdout",
@@ -56,7 +62,7 @@
     ")\n",
     "\n",
     "acc_predictor_checkpoint_path = download_url(\n",
-    "    'https://hanlab.mit.edu/files/OnceForAll/tutorial/ofa_resnet50_acc_predictor.pth',\n",
+    "    'https://raw.githubusercontent.com/han-cai/files/master/ofa/ofa_resnet50_acc_predictor.pth',\n",
     "    model_dir='~/.ofa/',\n",
     ")\n",
     "device = 'cuda:0' if torch.cuda.is_available() else 'cpu'\n",
@@ -65,34 +71,34 @@
     "\n",
     "print('The accuracy predictor is ready!')\n",
     "print(acc_predictor)"
-   ],
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
    "metadata": {
     "collapsed": false,
     "pycharm": {
      "name": "#%%\n"
     }
-   }
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 6,
+   },
    "outputs": [],
    "source": [
     "# build efficiency predictor\n",
     "from ofa.nas.efficiency_predictor import ResNet50FLOPsModel\n",
     "\n",
     "efficiency_predictor = ResNet50FLOPsModel(ofa_network)"
-   ],
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
    "metadata": {
     "collapsed": false,
     "pycharm": {
      "name": "#%%\n"
     }
-   }
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 11,
+   },
    "outputs": [
     {
      "name": "stdout",
@@ -123,13 +129,7 @@
     "    predicted_efficiency = efficiency_predictor.get_efficiency(subnet_config)\n",
     "\n",
     "    print(i, '\\t', predicted_acc, '\\t', '%.1fM MACs' % predicted_efficiency)\n"
-   ],
-   "metadata": {
-    "collapsed": false,
-    "pycharm": {
-     "name": "#%%\n"
-    }
-   }
+   ]
   }
  ],
  "metadata": {