model_tools: add input info to configs (#3183)

* support input info in configs

* update public models configs

* update intel models configs

Author: Julia Kamelina

CONTRIBUTING.md

@@ -113,6 +113,16 @@ For replacement operation:
 - `replacement` — Replacement string
 - `count` (*optional*)  — Exact number of replacements (if number of `pattern` occurrences less then this number, downloading will be aborted)
 
+**`input_info`**
+
+List of inputs containing the information about input name, shape and layout. For example:
+```
+input_info:
+  - name: Placeholder
+    shape: [1, 224, 224, 3]
+    layout: NHWC
+```
+
 **`conversion_to_onnx_args`** (*only for PyTorch\* models*)
 
 List of ONNX\* conversion parameters, see `model_optimizer_args` for details.
@@ -121,7 +131,6 @@ List of ONNX\* conversion parameters, see `model_optimizer_args` for details.
 
 Conversion parameters (learn more in the [Model conversion](#model-conversion) section). For example:
 ```
-  - --input=data
   - --mean_values=data[127.5]
   - --scale_values=data[127.5]
   - --reverse_input_channels
@@ -163,10 +172,12 @@ postprocessing:
   - $type: unpack_archive
     format: gztar
     file: tf-densenet121.tar.gz
+input_info:
+  - name: Placeholder
+    shape: [1, 224, 224, 3]
+    layout: NHWC
 model_optimizer_args:
   - --reverse_input_channels
-  - --input_shape=[1,224,224,3]
-  - --input=Placeholder
   - --mean_values=Placeholder[123.68,116.78,103.94]
   - --scale_values=Placeholder[58.8235294117647]
   - --output=densenet121/predictions/Reshape_1

models/intel/action-recognition-0001/action-recognition-0001-decoder/model.yml

@@ -33,5 +33,9 @@ files:
     size: 15119354
     checksum: dcf216af54725539aeeaddefd23d6511d475fecb472569377005e99ffd33f028ab7e98c0141a16f3704c5d0dae414e79
     source: https://storage.openvinotoolkit.org/repositories/open_model_zoo/2022.1/models_bin/2/action-recognition-0001/action-recognition-0001-decoder/FP16/action-recognition-0001-decoder.bin
+input_info:
+  - name: '0'
+    shape: [1, 16, 512]
+    layout: NDC
 framework: dldt
 license: https://raw.githubusercontent.com/openvinotoolkit/open_model_zoo/master/LICENSE

models/intel/action-recognition-0001/action-recognition-0001-encoder/model.yml

@@ -33,5 +33,9 @@ files:
     size: 42584582
     checksum: 5c0ac9530d7550e3a6d02a3f9bc3668d42804289530022c4419a661eb60e2ee507577728ad353287a94de99011a137ef
     source: https://storage.openvinotoolkit.org/repositories/open_model_zoo/2022.1/models_bin/2/action-recognition-0001/action-recognition-0001-encoder/FP16/action-recognition-0001-encoder.bin
+input_info:
+  - name: '0'
+    shape: [1, 3, 224, 224]
+    layout: NCHW
 framework: dldt
 license: https://raw.githubusercontent.com/openvinotoolkit/open_model_zoo/master/LICENSE

models/intel/age-gender-recognition-retail-0013/model.yml

@@ -40,5 +40,9 @@ files:
     size: 2158730
     checksum: 49dcdf7e7834ed88b4fc6a0ad6b8806febacd3d62f91cd9cdd8a461fac06557cb069c8045a5936c73553947091dcae05
     source: https://storage.openvinotoolkit.org/repositories/open_model_zoo/2022.1/models_bin/2/age-gender-recognition-retail-0013/FP16-INT8/age-gender-recognition-retail-0013.bin
+input_info:
+  - name: data
+    shape: [1, 3, 62, 62]
+    layout: NCHW
 framework: dldt
 license: https://raw.githubusercontent.com/openvinotoolkit/open_model_zoo/master/LICENSE

models/intel/asl-recognition-0004/model.yml

@@ -40,5 +40,9 @@ files:
     size: 4391006
     checksum: 6c64454d66ad7505b7702e230758f880b793712921897f5d0338c974b1a8b0d4e6b39dc6c00bb5d1bd95ee612b6e59f2
     source: https://storage.openvinotoolkit.org/repositories/open_model_zoo/2022.1/models_bin/2/asl-recognition-0004/FP16-INT8/asl-recognition-0004.bin
+input_info:
+  - name: input
+    shape: [1, 3, 16, 224, 224]
+    layout: NCDHW
 framework: dldt
 license: https://raw.githubusercontent.com/openvinotoolkit/open_model_zoo/master/LICENSE

models/intel/bert-large-uncased-whole-word-masking-squad-0001/model.yml

@@ -37,5 +37,15 @@ files:
     size: 668191930
     checksum: ec2365f5f3937b69b1a8424fe3b5e3db2389bcd6e31f4857f66cd385dccc23432a5c9d722f4ee278d2ba4b7b4f3b7a7d
     source: https://storage.openvinotoolkit.org/repositories/open_model_zoo/2022.1/models_bin/2/bert-large-uncased-whole-word-masking-squad-0001/FP16/bert-large-uncased-whole-word-masking-squad-0001.bin
+input_info:
+  - name: input_ids
+    shape: [1, 384]
+    layout: NC
+  - name: attention_mask
+    shape: [1, 384]
+    layout: NC
+  - name: token_type_ids
+    shape: [1, 384]
+    layout: NC
 framework: dldt
 license: https://raw.githubusercontent.com/openvinotoolkit/open_model_zoo/master/LICENSE

models/intel/bert-large-uncased-whole-word-masking-squad-emb-0001/model.yml

@@ -38,5 +38,18 @@ files:
     size: 670283942
     checksum: ed39d55a0a658cc3ca06d0bfb07fd03f7f76fcfd5fe59ab2e1c7357337c962270a7b064667b6a96b37ff3b203b0bbc53
     source: https://storage.openvinotoolkit.org/repositories/open_model_zoo/2022.1/models_bin/2/bert-large-uncased-whole-word-masking-squad-emb-0001/FP16/bert-large-uncased-whole-word-masking-squad-emb-0001.bin
+input_info:
+  - name: input_ids
+    shape: [1, 32]
+    layout: NC
+  - name: attention_mask
+    shape: [1, 32]
+    layout: NC
+  - name: token_type_ids
+    shape: [1, 32]
+    layout: NC
+  - name: position_ids
+    shape: [1, 32]
+    layout: NC
 framework: dldt
 license: https://raw.githubusercontent.com/openvinotoolkit/open_model_zoo/master/LICENSE

models/intel/bert-large-uncased-whole-word-masking-squad-int8-0001/model.yml

@@ -38,5 +38,15 @@ files:
     size: 366197980
     checksum: 8a12aa656586098b096bad6c95053ab0905ba70f70c472e61f4c3c745c2f9b9eb5fe19cb7bac4d040b63d8551d97f1e4
     source: https://storage.openvinotoolkit.org/repositories/open_model_zoo/2022.1/models_bin/2/bert-large-uncased-whole-word-masking-squad-int8-0001/FP16-INT8/bert-large-uncased-whole-word-masking-squad-int8-0001.bin
+input_info:
+  - name: input_ids
+    shape: [1, 384]
+    layout: NC
+  - name: attention_mask
+    shape: [1, 384]
+    layout: NC
+  - name: token_type_ids
+    shape: [1, 384]
+    layout: NC
 framework: dldt
 license: https://raw.githubusercontent.com/openvinotoolkit/open_model_zoo/master/LICENSE

models/intel/bert-small-uncased-whole-word-masking-squad-0001/model.yml

@@ -39,5 +39,15 @@ files:
     size: 116149434
     checksum: b3602c8f88b46bba4173a2f6e22e48161e5c0246628b148b70d1a84e930a8731447b93784c74ff20c5a6a0421768286f
     source: https://storage.openvinotoolkit.org/repositories/open_model_zoo/2022.1/models_bin/2/bert-small-uncased-whole-word-masking-squad-0001/FP16/bert-small-uncased-whole-word-masking-squad-0001.bin
+input_info:
+  - name: input_ids
+    shape: [1, 384]
+    layout: NC
+  - name: attention_mask
+    shape: [1, 384]
+    layout: NC
+  - name: token_type_ids
+    shape: [1, 384]
+    layout: NC
 framework: dldt
 license: https://raw.githubusercontent.com/openvinotoolkit/open_model_zoo/master/LICENSE

models/intel/bert-small-uncased-whole-word-masking-squad-0002/model.yml

@@ -39,5 +39,18 @@ files:
     size: 82264250
     checksum: 3c2bde8b280364834bc035836138ace202d090e483254186526229b3c37f4b9a23c1b655eaba5119468d4fa88a5de7bf
     source: https://storage.openvinotoolkit.org/repositories/open_model_zoo/2022.1/models_bin/2/bert-small-uncased-whole-word-masking-squad-0002/FP16/bert-small-uncased-whole-word-masking-squad-0002.bin
+input_info:
+  - name: input_ids
+    shape: [1, 384]
+    layout: NC
+  - name: attention_mask
+    shape: [1, 384]
+    layout: NC
+  - name: token_type_ids
+    shape: [1, 384]
+    layout: NC
+  - name: position_ids
+    shape: [1, 384]
+    layout: NC
 framework: dldt
 license: https://raw.githubusercontent.com/openvinotoolkit/open_model_zoo/master/LICENSE