!9794 [MSLITE]fix the error that occurs when the denominator of Rsprt is 0

From: @probiotics_53
Reviewed-by: @zhang_xue_tong,@hangangqiang
Signed-off-by: @hangangqiang

mindspore/lite/nnacl/errorcode.h

@@ -30,6 +30,7 @@ typedef enum ErrorCodeFp32OpEnum {
   NNACL_ERRCODE_STRASSEN_RECURSION_MALLOC,
   NNACL_ERRCODE_REVERSE_MALLOC,
   NNACL_ERRCODE_SQRT_NEGATIVE,
+  NNACL_ERRCODE_RSQRT_NEGATIVE,
   NNACL_ERRCODE_RSQRT_NEGATIVE_OR_ZERO,
   NNACL_ERRCODE_LOG_NEGATIVE_OR_ZERO,
   NNACL_ERRCODE_DIVISOR_ZERO,

mindspore/lite/nnacl/fp32/arithmetic_self_fp32.c

@@ -68,8 +68,8 @@ int ElementSqrt(const float *input, float *output, const int element_size) {
 // rsqrt
 int ElementRsqrt(const float *input, float *output, const int element_size) {
   for (int i = 0; i < element_size; i++) {
-    if (input[i] <= 0) {
-      return NNACL_ERRCODE_RSQRT_NEGATIVE_OR_ZERO;
+    if (input[i] < 0) {
+      return NNACL_ERRCODE_RSQRT_NEGATIVE;
     }
     output[i] = 1.f / sqrtf(input[i]);
   }

mindspore/lite/nnacl/int8/quant_dtype_cast_int8.c

@@ -36,10 +36,14 @@ int DoQuantizeFp32ToInt8(const float *real_values, int8_t *quant_values, float s
 
   const float inverse_scale = 1.0f / scale;
   for (int i = 0; i < size; ++i) {
-    int temp = round(real_values[i] * inverse_scale + zp);
-    temp = temp < 127 ? temp : 127;
-    temp = temp > -128 ? temp : -128;
-    quant_values[i] = (int8_t)temp;
+    if (isinf(real_values[i])) {
+      quant_values[i] = 127;
+    } else {
+      int temp = round(real_values[i] * inverse_scale + zp);
+      temp = temp < 127 ? temp : 127;
+      temp = temp > -128 ? temp : -128;
+      quant_values[i] = (int8_t)temp;
+    }
   }
   return NNACL_OK;
 }
@@ -61,13 +65,17 @@ int DoQuantizeFp32ToUInt8(const float *real_values, uint8_t *quant_values, float
   }
 
   for (int i = 0; i < size; ++i) {
-    float temp = (float)round(real_values[i] * 1.0 / scale + zp);
-    if (temp > 255) {
+    if (isinf(real_values[i])) {
       quant_values[i] = 255;
-    } else if (temp < 0) {
-      quant_values[i] = 0;
     } else {
-      quant_values[i] = (uint8_t)temp;
+      float temp = (float)round(real_values[i] * 1.0 / scale + zp);
+      if (temp > 255) {
+        quant_values[i] = 255;
+      } else if (temp < 0) {
+        quant_values[i] = 0;
+      } else {
+        quant_values[i] = (uint8_t)temp;
+      }
     }
   }
   return NNACL_OK;

mindspore/lite/tools/benchmark/benchmark.cc

@@ -443,6 +443,8 @@ int Benchmark::PrintInputData() {
         std::cout << static_cast<const uint8_t *>(in_data)[j] << " ";
       } else if (tensor_data_type == TypeId::kNumberTypeInt32) {
         std::cout << static_cast<const int32_t *>(in_data)[j] << " ";
+      } else if (tensor_data_type == TypeId::kNumberTypeInt64) {
+        std::cout << static_cast<const int64_t *>(in_data)[j] << " ";
       } else {
         MS_LOG(ERROR) << "Datatype: " << tensor_data_type << " is not supported.";
         return RET_ERROR;

mindspore/lite/tools/converter/parser/tflite/tflite_dequantize_parser.cc

@@ -33,8 +33,7 @@ PrimitiveC *TfliteDequantizeParser::ParseLitePrimitive(const std::unique_ptr<tfl
     MS_LOG(ERROR) << "output tensor is null";
     return nullptr;
   }
-  if (GetTfliteDataType(in_tensor->type) != GetTfliteDataType(out_tensor->type) &&
-      (GetTfliteDataType(in_tensor->type) == kNumberTypeInt8 ||
+  if ((GetTfliteDataType(in_tensor->type) == kNumberTypeInt8 ||
        GetTfliteDataType(in_tensor->type) == kNumberTypeUInt8)) {
     std::unique_ptr<schema::QuantDTypeCastT> attr = std::make_unique<schema::QuantDTypeCastT>();
     if (attr == nullptr) {

mindspore/lite/tools/converter/parser/tflite/tflite_quantize_parser.cc

@@ -38,8 +38,7 @@ PrimitiveC *TfliteQuantizeParser::ParseLitePrimitive(const std::unique_ptr<tflit
     MS_LOG(ERROR) << "output tensor is null";
     return nullptr;
   }
-  if (GetTfliteDataType(in_tensor->type) != GetTfliteDataType(out_tensor->type) &&
-      (GetTfliteDataType(out_tensor->type) == kNumberTypeInt8 ||
+  if ((GetTfliteDataType(out_tensor->type) == kNumberTypeInt8 ||
        GetTfliteDataType(out_tensor->type) == kNumberTypeUInt8)) {
     std::unique_ptr<schema::QuantDTypeCastT> attr = std::make_unique<schema::QuantDTypeCastT>();
     if (attr == nullptr) {