remove activation quantization of slb

official/cv/resnet/README.md

@@ -892,28 +892,22 @@ result:{'top_1_accuracy': 0.9354967948717948, 'top_5_accuracy': 0.99819711538461
 result:{'top_1_accuracy': 0.9273838141025641} prune_rate=0.45 ckpt=~/resnet50_cifar10/train_parallel0/resnet-400_390.ckpt
 ```
 
-- Apply SLB on ResNet18 with W4A8, and evaluating with CIFAR-10 dataset. W4A8 means quantize weight with 4bit and activation with 8bit:
+- Apply SLB on ResNet18 with W4, and evaluating with CIFAR-10 dataset. W4 means quantize weight with 4bit:
 
 ```text
-result:{'top_1_accuracy': 0.9285857371794872, 'top_5_accuracy': 0.9959935897435898} ckpt=~/resnet18_cifar10/train_parallel/resnet-100_1562.ckpt
+result:{'top_1_accuracy': 0.9534254807692307, 'top_5_accuracy': 0.9969951923076923} ckpt=~/resnet18_cifar10/train_parallel/resnet-100_1562.ckpt
 ```
 
-- Apply SLB on ResNet18 with W2A8, and evaluating with CIFAR-10 dataset. W2A8 means quantize weight with 2bit and activation with 8bit:
+- Apply SLB on ResNet18 with W2, and evaluating with CIFAR-10 dataset. W2 means quantize weight with 2bit:
 
 ```text
-result:{'top_1_accuracy': 0.9207732371794872, 'top_5_accuracy': 0.9955929487179487} ckpt=~/resnet18_cifar10/train_parallel/resnet-100_1562.ckpt
+result:{'top_1_accuracy': 0.9503205128205128, 'top_5_accuracy': 0.9966947115384616} ckpt=~/resnet18_cifar10/train_parallel/resnet-100_1562.ckpt
 ```
 
-- Apply SLB on ResNet18 with W1A8, and evaluating with CIFAR-10 dataset. W1A8 means quantize weight with 1bit and activation with 8bit:
+- Apply SLB on ResNet18 with W1, and evaluating with CIFAR-10 dataset. W1 means quantize weight with 1bit:
 
 ```text
-result:{'top_1_accuracy': 0.8976362179487182, 'top_5_accuracy': 0.9923878205128205} ckpt=~/resnet18_cifar10/train_parallel/resnet-100_1562.ckpt
-```
-
-- Apply SLB on ResNet18 with W1A4, and evaluating with CIFAR-10 dataset. W1A4 means quantize weight with 1bit and activation with 4bit:
-
-```text
-result:{'top_1_accuracy': 0.8845152243589743, 'top_5_accuracy': 0.9914863782051282} ckpt=~/resnet18_cifar10/train_parallel/resnet-100_1562.ckpt
+result:{'top_1_accuracy': 0.9485176282051282, 'top_5_accuracy': 0.9965945512820513} ckpt=~/resnet18_cifar10/train_parallel/resnet-100_1562.ckpt
 ```
 
 # [Model Description](#contents)

official/cv/resnet/README_CN.md

@@ -720,7 +720,7 @@ bash run_infer_310.sh [MINDIR_PATH] [NET_TYPE] [DATASET] [DATA_PATH] [CONFIG_PAT
 - 使用CIFAR-10数据集评估ResNet18
 
 ```bash
-Total data: 10000, top1 accuracy: 0.94.26, top5 accuracy: 0.9987.
+Total data: 10000, top1 accuracy: 0.9426, top5 accuracy: 0.9987.
 ```
 
 - 使用ImageNet2012数据集评估ResNet18
@@ -854,28 +854,22 @@ result:{'top_1_accuracy': 0.9354967948717948, 'top_5_accuracy': 0.99819711538461
 result:{'top_1_accuracy': 0.9273838141025641} prune_rate=0.45 ckpt=~/resnet50_cifar10/train_parallel0/resnet-400_390.ckpt
 ```
 
-- 使用SLB算法对ResNet18做W4A8量化，并使用CIFAR-10数据集评估，W4A8表示weight量化为4bit，activation量化为8bit：
+- 使用SLB算法对ResNet18做W4量化，并使用CIFAR-10数据集评估，W4表示weight量化为4bit：
 
 ```text
-result:{'top_1_accuracy': 0.9285857371794872, 'top_5_accuracy': 0.9959935897435898} ckpt=~/resnet18_cifar10/train_parallel/resnet-100_1562.ckpt
+result:{'top_1_accuracy': 0.9534254807692307, 'top_5_accuracy': 0.9969951923076923} ckpt=~/resnet18_cifar10/train_parallel/resnet-100_1562.ckpt
 ```
 
-- 使用SLB算法对ResNet18做W2A8量化，并使用CIFAR-10数据集评估，W2A8表示weight量化为2bit，activation量化为8bit：
+- 使用SLB算法对ResNet18做W2量化，并使用CIFAR-10数据集评估，W2表示weight量化为2bit：
 
 ```text
-result:{'top_1_accuracy': 0.9207732371794872, 'top_5_accuracy': 0.9955929487179487} ckpt=~/resnet18_cifar10/train_parallel/resnet-100_1562.ckpt
+result:{'top_1_accuracy': 0.9503205128205128, 'top_5_accuracy': 0.9966947115384616} ckpt=~/resnet18_cifar10/train_parallel/resnet-100_1562.ckpt
 ```
 
-- 使用SLB算法对ResNet18做W1A8量化，并使用CIFAR-10数据集评估，W1A8表示weight量化为1bit，activation量化为8bit：
+- 使用SLB算法对ResNet18做W1量化，并使用CIFAR-10数据集评估，W1表示weight量化为1bit：
 
 ```text
-result:{'top_1_accuracy': 0.8976362179487182, 'top_5_accuracy': 0.9923878205128205} ckpt=~/resnet18_cifar10/train_parallel/resnet-100_1562.ckpt
-```
-
-- 使用SLB算法对ResNet18做W1A4量化，并使用CIFAR-10数据集评估，W1A4表示weight量化为1bit，activation量化为4bit：
-
-```text
-result:{'top_1_accuracy': 0.8845152243589743, 'top_5_accuracy': 0.9914863782051282} ckpt=~/resnet18_cifar10/train_parallel/resnet-100_1562.ckpt
+result:{'top_1_accuracy': 0.9485176282051282, 'top_5_accuracy': 0.9965945512820513} ckpt=~/resnet18_cifar10/train_parallel/resnet-100_1562.ckpt
 ```
 
 # 模型描述

official/cv/resnet/golden_stick/quantization/slb/eval.py

@@ -13,6 +13,7 @@
 # limitations under the License.
 # ============================================================================
 """eval resnet."""
+
 import mindspore as ms
 import mindspore.log as logger
 from mindspore.nn.loss import SoftmaxCrossEntropyWithLogits

official/cv/resnet/golden_stick/quantization/slb/resnet18_cifar10_config.yaml

@@ -60,7 +60,7 @@ eval_image_size: 224
 
 # Golden-stick options
 comp_algo: "SLB"
-quant_type: "W1A8"
+quant_type: "W1"
 t_start_val: 1.0
 t_start_time: 0.2
 t_end_time: 0.6

official/cv/resnet/golden_stick/quantization/slb/slb.py

@@ -14,21 +14,15 @@
 # ============================================================================
 """Create SLB-QAT algorithm instance."""
 
-from mindspore_gs.quantization.slb import SlbQuantAwareTraining as QBNNQAT
+from mindspore_gs.quantization.slb import SlbQuantAwareTraining as SlbQAT
 from mindspore_gs.quantization.constant import QuantDtype
 
-def create_slb(quant_type="graph_W4A8"):
-    algo = QBNNQAT()
-    if "W4A8" in quant_type:
+def create_slb(quant_type="W1"):
+    algo = SlbQAT()
+    if "W4" in quant_type:
         algo.set_weight_quant_dtype(QuantDtype.INT4)
-        algo.set_act_quant_dtype(QuantDtype.INT8)
-    elif "W2A8" in quant_type:
+    elif "W2" in quant_type:
         algo.set_weight_quant_dtype(QuantDtype.INT2)
-        algo.set_act_quant_dtype(QuantDtype.INT8)
-    elif "W1A8" in quant_type:
+    elif "W1" in quant_type:
         algo.set_weight_quant_dtype(QuantDtype.INT1)
-        algo.set_act_quant_dtype(QuantDtype.INT8)
-    elif "W1A4" in quant_type:
-        algo.set_weight_quant_dtype(QuantDtype.INT1)
-        algo.set_act_quant_dtype(QuantDtype.INT4)
     return algo

official/cv/resnet/golden_stick/quantization/slb/train.py

@@ -13,9 +13,9 @@
 # limitations under the License.
 # ============================================================================
 """train resnet."""
+
 import os
 import numpy as np
-
 import mindspore as ms
 import mindspore.nn as nn
 import mindspore.train.callback as callback
@@ -130,7 +130,7 @@ def load_pretrained_ckpt(net):
     if config.pre_trained:
         if os.path.isfile(config.pre_trained):
             ckpt = ms.load_checkpoint(config.pre_trained)
-            if False and ckpt.get("epoch_num") and ckpt.get("step_num"):
+            if ckpt.get("epoch_num") and ckpt.get("step_num"):
                 config.has_trained_epoch = int(ckpt["epoch_num"].data.asnumpy())
                 config.has_trained_step = int(ckpt["step_num"].data.asnumpy())
             else:
@@ -208,7 +208,7 @@ class TemperatureScheduler(callback.Callback):
             t *= self.t_factor**(min(epoch, t_end_epoch) - t_start_epoch)
         # Assign new value to temperature parameter
         for _, cell in self.model.train_network.cells_and_names():
-            if cell.cls_name == 'QBNNFakeQuantizerPerLayer': # for QBNN
+            if cell.cls_name == 'SlbFakeQuantizerPerLayer': # for SLB
                 cell.set_temperature(t)
                 if epoch >= t_end_epoch:
                     cell.set_temperature_end_flag()