update docs

examples/FasterRCNN/README.md

@@ -54,12 +54,12 @@ MaskRCNN results contain both bbox and segm mAP.
 |Backbone | `FASTRCNN_BATCH` | resolution | mAP (bbox/segm) | Time |
 | - | - | - | - | - |
 | Res50 | 64 | (600, 1024) | 33.0 | 22h on 8 P100 |
-| Res50 | 256 | (600, 1024) | 34.4 | 49h on 8 TitanX |
+| Res50 | 256 | (600, 1024) | 34.4 | 49h on 8 M40 |
 | Res50 | 512 | (800, 1333) | 35.6 | 55h on 8 P100|
 | Res50 | 512 | (800, 1333) | 36.9/32.3 | 59h on 8 P100|
 | Res101 | 512 | (800, 1333) | 40.1/34.4 | 70h on 8 P100|
 
-Note that these models are trained with a longer learning schedule than the paper,
+Note that these models are trained with a larger ROI batch size than the paper,
 and get about 1mAP better performance.
 
 ## Notes

examples/FasterRCNN/config.py

@@ -35,7 +35,7 @@ NEGATIVE_ANCHOR_THRES = 0.3
 BBOX_DECODE_CLIP = np.log(MAX_SIZE / 16.0)
 
 # rpn training -------------------------
-# keep fg ratio in a batch in this range
+# fg ratio among selected RPN anchors
 RPN_FG_RATIO = 0.5
 RPN_BATCH_PER_IM = 256
 RPN_MIN_SIZE = 0
@@ -49,7 +49,7 @@ CROWD_OVERLAP_THRES = 0.7
 FASTRCNN_BATCH_PER_IM = 256
 FASTRCNN_BBOX_REG_WEIGHTS = np.array([10, 10, 5, 5], dtype='float32')
 FASTRCNN_FG_THRESH = 0.5
-# keep fg ratio in a batch in this range
+# fg ratio in a ROI batch
 FASTRCNN_FG_RATIO = 0.25
 
 # testing -----------------------