Merge branch 'main' into channels_last_branches

README.md

@@ -101,6 +101,7 @@ Inference cost for CNV_2W2A.onnx
 }
 ```
 
+You can use the `--cost-breakdown` option to generate a more detailed report that covers per-node (by name) and per-op-type information.
 You can read more about the BOPS metric in [this paper](https://www.frontiersin.org/articles/10.3389/frai.2021.676564/full), Section 4.2 Bit Operations.
 
 ### Convert between different quantization representations
@@ -114,16 +115,19 @@ Please see the documentation of the `QuantToQCDQ` transformation to learn more a
 
 ## Development
 
-Install in editable mode in a venv:
+Install in editable mode in a Python virtual environment:
 
 ```
 git clone https://github.com/fastmachinelearning/qonnx
 cd qonnx
 virtualenv -p python3.8 venv
 source venv/bin/activate
+pip install --upgrade pip
 pip install -e .[qkeras,testing]
 ```
 
+### Running tests
+
 Run entire test suite, parallelized across CPU cores:
 ```
 pytest -n auto --verbose
@@ -134,6 +138,22 @@ Run a particular test and fall into pdb if it fails:
 pytest --pdb -k "test_extend_partition.py::test_extend_partition[extend_id1-2]"
 ```
 
+### Linting
+
+If you plan to make pull requests to the qonnx repo, linting will be required.
+We use a pre-commit hook to auto-format Python code and check for issues. See https://pre-commit.com/ for installation. Once you have `pre-commit`,
+you can install the hooks into your local clone of the qonnx repo:
+
+```
+cd qonnx
+source venv/bin/activate
+pip install pre-commit
+pre-commit install
+```
+
+Every time you commit some code, the pre-commit hooks will first run, performing various checks and fixes. In some cases pre-commit won’t be able to
+fix the issues and you may have to fix it manually, then run git commit once again. The checks are configured in .pre-commit-config.yaml under the repo root.
+
 ## Why QONNX?
 
 The QONNX representation has several advantages compared to other alternatives, as summarized in the table below.

docs/index.rst

@@ -63,6 +63,9 @@ Install in editable mode in a venv:
    pip install -e .[testing, docs, notebooks]
 
 
+Test suite
+++++++++++
+
 Run entire test suite, parallelized across CPU cores:
 
 ::

docs/license.rst

@@ -1,7 +1,7 @@
 .. _license:
 
-=======
+========
 License
-=======
+========
 
 .. include:: ../LICENSE

docs/qonnx-custom-ops/quant_op.md

@@ -21,7 +21,7 @@ This operator is not part of the ONNX standard and is not currently versioned.
 <dt><tt>narrow</tt> : int (default is 0)</dt>
 <dd>Defines if the value range should be interpreted as narrow, when signed=1. E.g. at 8b regular=[-128, 127] vs narrow=[-127, 127].</dd>
 <dt><tt>rounding_mode</tt> : string (default is "ROUND")</dt>
-<dd>Defines how rounding should be applied during quantization. Currently available modes are: "ROUND", "CEIL" and "FLOOR". Here "ROUND" implies a round-to-even operation. Lowercase variants for the rounding mode string are also supported: "round", "ceil", "floor".</dd>
+<dd>Defines how rounding should be applied during quantization. Avaiable options are ROUND, CEIL, FLOOR, UP, DOWN, HALF_UP, HALF_DOWN. The rounding modes are described in the table bellow. The names of rounding modes can be upper case or lower case.</dd>
 </dl>
 
 #### Inputs
@@ -46,6 +46,24 @@ This operator is not part of the ONNX standard and is not currently versioned.
 </dl>
 
 
+#### Rounding modes
+<details>
+<summary>rounding modes</summary>
+
+| **Number \ ROUNDING_MODE** | ROUND=HALF_EVEN | CEIL | FLOOR | UP | DOWN | HALF_UP | HALF_DOWN |
+|----------------------------|-----------------|------|-------|----|------|---------|-----------|
+| 5.5                        | 6               | 6    | 5     | 6  | 5    | 6       | 5         |
+| 2.5                        | 2               | 3    | 2     | 3  | 2    | 3       | 2         |
+| 1.6                        | 2               | 2    | 1     | 2  | 1    | 2       | 2         |
+| 1.1                        | 1               | 2    | 1     | 2  | 1    | 1       | 1         |
+| 1.0                        | 1               | 1    | 1     | 1  | 1    | 1       | 1         |
+| -1.0                       | -1              | -1   | -1    | -1 | -1   | -1      | -1        |
+| -1.1                       | -1              | -1   | -2    | -2 | -1   | -1      | -1        |
+| -1.6                       | -2              | -1   | -2    | -2 | -1   | -2      | -2        |
+| -2.5                       | -2              | -2   | -3    | -3 | -2   | -3      | -2        |
+| -5.5                       | -6              | -5   | -6    | -6 | -5   | -6      | -5        |
+</details>
+
 #### Examples
 <details>
 <summary>Quant</summary>

notebooks/0_how_to_work_with_onnx.ipynb

@@ -68,26 +68,26 @@
     "    name='Add2',\n",
     ")\n",
     "\n",
-    "Add3_node = onnx.helper.make_node(\n",
-    "    'Add',\n",
-    "    inputs=['abs1', 'abs1'],\n",
-    "    outputs=['sum3'],\n",
-    "    name='Add3',\n",
-    ")\n",
-    "\n",
     "Abs_node = onnx.helper.make_node(\n",
     "    'Abs',\n",
     "    inputs=['sum2'],\n",
     "    outputs=['abs1'],\n",
     "    name='Abs'\n",
     ")\n",
     "\n",
+    "Add3_node = onnx.helper.make_node(\n",
+    "    'Add',\n",
+    "    inputs=['abs1', 'abs1'],\n",
+    "    outputs=['sum3'],\n",
+    "    name='Add3',\n",
+    ")\n",
+    "\n",
     "Round_node = onnx.helper.make_node(\n",
     "    'Round',\n",
     "    inputs=['sum3'],\n",
     "    outputs=['out1'],\n",
     "    name='Round',\n",
-    ")\n"
+    ")"
    ]
   },
   {
@@ -253,7 +253,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "in1_values =np.asarray(np.random.uniform(low=-5, high=5, size=(4,4)), dtype=np.float32)\n",
+    "in1_values = np.asarray(np.random.uniform(low=-5, high=5, size=(4,4)), dtype=np.float32)\n",
     "in2_values = np.asarray(np.random.uniform(low=-5, high=5, size=(4,4)), dtype=np.float32)\n",
     "in3_values = np.asarray(np.random.uniform(low=-5, high=5, size=(4,4)), dtype=np.float32)"
    ]
@@ -350,6 +350,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "import qonnx\n",
     "from qonnx.core.modelwrapper import ModelWrapper\n",
     "finn_model = ModelWrapper(onnx_model)"
    ]